Jean Calterone Williams, Domestic Violence and Poverty:  The Narratives of Homeless Women, 19 Frontiers 143 (1998).

Among the many reasons for homelessness, domestic violence and low-cost
housing shortages experienced within a context of poverty are fundamental
for low income women living in shelters. Women interviewed in homeless and
battered women's shelters in Phoenix, Arizona, describe a process of becoming
homeless that usually involves a combination of interlocking events and
factors, such that it is impossible to isolate one explanation for a woman's
homelessness. Nonetheless, women's stories indicate a pattern in their
persistent poverty and battering relationships prior to becoming homeless.


Most research distinguishes between the women who live in homeless shelters
and those in domestic violence shelters. Likewise, the environments and
programs offered by the two types of shelters vary significantly, and shelter
staff usually argue that a homeless woman has different issues and needs
than a battered woman. This study concurrently analyzes both kinds of shelters
and points to the striking similarities in women's reasons for seeking
emergency housing whether they reside in domestic violence or homeless
shelters. Specifically, women discuss similarly impoverished circumstances
and often indicate that their past histories include abusive partners.
Partly, the similarity in women's stories can be traced to the overlap
in populations at the two types of shelters. A woman who has left an abusive
relationship may enter a homeless shelter after spending thirty days at
a domestic violence shelter. Shifting to a homeless shelter is in part
the effect of battered women's shelter policies, which encourage women
to look for other accommodations after thirty days have expired. Other
women may go directly from their relationships to homeless shelters because
domestic violence shelters are full. Others prefer the attention paid to
economic issues or even the generally later curfews found at homeless shelters.
The overlap in client populations in the two kinds of shelters and the
similarities among women's stories-in particular, the centrality of battering-suggest
a complex connection between battering and homelessness. Such a connection
is rarely captured by the easy distinctions generally drawn between women
who are identified by shelter staff either as battered or homeless.

The distinctions drawn between battered women and homeless women by shelters
is reflected in academic and popular literature. Though some homelessness
studies address domestic violence as a central concern,1 such violence
is more often mentioned only in passing as a reason for women's homelessness,
and most studies neither explore its significance for homeless women nor
examine the process of becoming homeless as it results from domestic violence.
For their part, analyses of domestic violence take more seriously the link
between women's financial status and their ability to resist violence,2
but only a few seriously explore homelessness.3

Though much attention has been paid to the "feminization of poverty," there
has been less emphasis on the significance of poverty for increasing the
likelihood that women will become homeless.4 Female single-parent families
rose from 23.7 percent of all families in poverty in 1960 to 52.6 percent
of all families in poverty in the mid-1990s.5 According to the 1995 U.S.
Conference of Mayor's report on homelessness, the poverty of female-headed
families is reflected in the composition of the homeless population. Of
the twenty-nine cities surveyed for the report, twenty-three reported that
at least 70 percent of homeless families in their cities were single-parent
families, with nineteen citing from 80 percent to 98 percent single-parent
families,6 the majority of them female headed. As a proportion of the homeless
population, families grew from 27 percent in 1985 to 43 percent in 1993
and dropped slightly to 38 percent in 1996.7 Moreover, seventeen of the
twenty-four cities that responded to a set of questions from the U.S. Conference
of Mayors regarding requests for shelter by families indicated an increase
in requests. Some cities reflected as much as a 100 percent increase from
1994 to 1995. As a result of historical growth in women's poverty and female-headed
family homelessness, it has become increasingly important for research
to focus on the unique sets of issues and problems that women's homelessness
presents.

Women interviewed for this study emphasize the impact of divorce, battering,
and other family disruptions in combination with economic insecurity and
primary responsibility for their children on their paths to homelessness.
Notwithstanding the complex story each woman tells when asked what events
led her to seek temporary shelter, the combination of persistent poverty,
domestic violence, and low-rent housing shortages were most often cited
as causing crises. Just as social ethnographers have discovered in interviews
with people receiving government assistance,8 most women and their families
managed to survive for some time with little income and precarious housing,
in other words, in an uncertain financial situation. However, when financial
difficulties occurred-problems like a car breakdown that would not cripple
a middle-income familythere was no surplus money to finance the unexpected.
Many of the women's parents and circle of friends were low-income as well,
but had not themselves experienced homelessness. Nevertheless they lacked
the resources needed to assist their homeless relatives with substantial
amounts of money or long-term housing.

Recent changes in the availability of welfare benefits through Aid to Families
with Dependent Children (AFDC) will also have an impact on women currently
living in shelters or those who are borderline homeless. Low-paid or seasonally
employed workers' loss of welfare benefits, however meager, may create
more homelessness as people are unable to meet rent, utility, and food
expenses during bouts of unemployment. The current services and shelters
designed to assist homeless people will likely be overwhelmed by an increase
in the numbers of the homeless. In fact, surveys of homeless shelters have
found significant numbers of homeless families and individuals already
are unable to find shelter due to a lack of space. Families often comprise
a disproportionate number of those unable to find a shelter with an opening,
as they are in Maryland, where families and children comprised 57 percent
of all people turned away in 1990 and about 50 percent in 1991, despite
making up closer to 30 or 40 percent of the homeless population nationwide.9


Methodology

This study is based on research that was conducted in homeless and domestic
violence shelters in Phoenix, Arizona, from 1994 to 1996. The research
included in-depth, semistructured interviews with thirty-three homeless
women and participant-observation in several homeless and battered women's
shelters with approximately one hundred shelter residents and the case
workers, usually professional social workers, hired to work with them.lo
A review of fifty randomly chosen 1995 case files at one family homeless
shelter supplements the interviews and participant-observation. In addition,
I worked for a year as a part-time caseworker at Rose's House, a domestic
violence shelter.

Interview questions were open-ended and as general as possible, encouraging
women to formulate their own narratives of homelessness. In this sense,
interview participants could include the information that to their minds
explains homelessness and delete parts of their life histories that they
believe do not count as an explanation of why someone loses housing. Their
stories are, of course, mediated by the author to the extent that I asked
certain questions and not others, and have the power to frame their stories
in a specific context. Though some women probably left out embarrassing
or incriminating information about themselves, most who discussed violence,
drug use, and sex work offered this information without direct questioning,
usually in response to an initial, general question regarding how they
became homeless.

In addition to relaying homeless women's lived experiences, this project
offers a story about the meanings attached to homelessness and domestic
violence, a story that has been shaped through interviews and informal
talks with homeless women. These interviews are rich in paradox and contradiction,
indicating a great diversity in personal experience. With such diversity,
I do not claim to offer the "homeless point of view"; rather, these voices
are multiple and conflicting.11 Yet they ultimately yield a story that
gives significance both to homelessness and domestic violence that helps
to make sense of their cultural and social meanings in the United States.


Most of the women interviewed came from two fairly small family homeless
shelters, The Family Shelter and Lighthouse, and two domestic violence
shelters for women and children, Rose's House and Casa Para Las Mujeres
(La Casa). These shelters vary in size, housing from five to thirty families.
All have cumbersome sets of rules and a significant amount of client and
staff contact. In contrast to larger or armory-style homeless shelters,
which tend to have fewer caseworkers per client and fewer expectations
of residents, the smaller, more tightly controlled shelters offer more
privacy and amenities, and fewer clients per caseworker. Although the more
comfortable living environment is accompanied by more stringent regulations,
these shelters are almost always full and consistently have to turn away
people because of a lack of space. Staff at the smaller family homeless
shelters tend to accept those who appear most "motivated" or most likely
to secure housing within the three months that they are allowed to live
at most shelters.l2 This practice usually excludes those who are severely
mentally ill and heavy substance users because they are probably least
likely to have the ability or desire to obey stricter shelter rules.13


Shelter Organization

As argued above, the interviews completed for this study point to the difficulties
in citing the one reason women turn to domestic violence and homeless shelters.
But analyses of homelessness often divide shelter residents and other homeless
people into groups of those who have lost housing as a result of drug use,
mental illness, or the vagaries of the economy. Although categorization
is usually offered as the first step in calculating how many people are
homeless and why, and how government and private agencies can best respond
to them,14 Peter Marcuse criticizes the practice. He argues that it is
symptomatic of a reliance on "specialism, or calling a general problem
the sum of a number of different special problems."15 Rather than focusing
on the systemic structural roots of homelessness, specialism both reflects
and encourages a cultural understanding of homelessness as merely a "mental
health, substance abuse, or criminal justice problem."16 The tendency of
such research to categorize people based on the one reason they lost housing
precludes a focus on the process of becoming homeless, on how, for example,
domestic violence, drug use, and poverty actually intertwine to lead to
homelessness.

Much like analyses of homelessness, each emergency shelter tends to direct
its services toward a certain subpopulation of the homeless. Abused and
neglected teens, drug addicts, homeless families, and battered women are
each assisted in a different program, to the exclusion of others who are
similarly without housing. Most shelters do not offer services to women
unless they conform to the criteria that ostensibly makes them part of
the particular group the shelter program targets. These classifications
are delimited by a range of "experiences and characteristics" that do not
necessarily match the women's lives who wish to gain entrance into a program.
17 As a result of their reliance on such categories, programs geared toward
battered and homeless women vary considerably in the importance given to
housing and employment considerations, lifestyle and behavioral issues,
and emotional state.

Women who do not clearly fit into either the battered woman or homeless
woman identity have great difficulty locating a program to assist them,
or may redefine their experiences in order to correspond more closely to
an identity looked for by a shelter. For example, a woman who is defined
by staff at a battered women's shelter as "streetwise" (often a euphemism
for "homeless" or "drug user") will have difficulty gaining entrance into
that shelter, whether or not she has left an abusive partner. Likewise,
some homeless shelters refuse to accept women who become homeless when
they leave abusive relationships, even if battered women's shelters are
full, because they do not define these women as "homeless." As the women's
stories that follow make clear, however, many women have multiple issues
for which they seek assistance but learn to emphasize one problem and to
conceal another in order to gain acceptance into a particular shelter.
Women's stories in this study indicate that the distinctions between "battered
woman" and "homeless woman" are not as straightforward nor absolute as
they often appear in cultural narratives drawn upon in the academic literature,
by shelter employees, and sometimes by shelter residents themselves.

Although they know all too well the financial constraints that have colored
their own decisions, battered women nevertheless often support and further
construct a differentiation between homeless and battered women apparent
in the emphases at different shelters. Unlike those who are in homeless
shelters, "battered women" can rely on an identity that names them "victims"
or perhaps "survivors," identities that do not hold them responsible for
living in a shelter and assumes a supportive community of women in similar
circumstances. Analyses of battering have noted that "women who have lived
through such violence who know the immense daily expenditure of strength
and attention and selfdiscipline it takes to survive, rarely identify themselves
as victims.' They think of themselves as strong women who can somehow `cope."''8
Though it could be as easily argued that the women in homeless shelters
develop personal fortitude through their struggles to survive, they tend
not to be constructed as "survivors." Culturally, "battered women" are
congratulated for turning to a shelter, as leaving their relationships
is constructed as the defining act of "helping themselves," while those
in homeless shelters generally are constructed as lazy and having turned
to shelters as a way to "live off the system."

Phyllis, a fifty-five-year-old Euro-American resident of Rose's House,
provides an example of the ways in which domestic violence shelter residents
distinguish themselves from "homeless women." She contends that homeless
people seek shelter after they have allowed their financial and personal
lives to completely deteriorate, while battered women are in shelters because
they want to change their lives, which have been ruined by someone else:
"I think I just ended up in this place through no fault of my own.... I've
found so much caring here, we really support each other, we have that sister-love."
Kim, a Euro-American woman in her mid-forties and also a resident of Rose's
House, responds similarly to a question about whether she considers herself
homeless:

No, not really. I've always been very resourceful.... I've got friends
who are battered women and come to shelters like this, but I don't know
anyone who's been homeless. We're survivors, most of us. We've learned
to be very strong and resourceful because of everything we've been through.
Financially, it's a big hurdle. If you don't have a job or skills, then
you've got a problem.

Despite cultural narratives of domestic violence that historically have
constructed battered women as "masochists,"'9 and a more current concentration
on the psychologically debilitating effects of battering,20 Kim represents
women who view leaving as courageous. Indeed, she argues that battered
women are stronger than most specifically because they have experienced
such violence. Tammy, also a EuroAmerican resident of Rose's House in her
early forties, echoes Kim's construction of battered women as survivors.
In response to the same question asked of Kim regarding whether she thinks
of herself as homeless, Tammy asserts: "No, and it's bizarre because I
guess I have been [homeless], but when someone says homeless you just go
'yuck'-I think of a bag lady. I guess since I've always been able to provide
for myself, I just think of myself as in transition instead of homeless."


Those women who become homeless for reasons other than domestic violence
display a similar desire to separate themselves from the label "homeless."
"Homeless" suggests a woman mentally ill or drug addicted, unwashed, helpless,
and hopeless, a person mired in a permanent "lifestyle" rather than "in
transition." Though the inclination remains as strong for homeless women
who are not battered, battered women make the move to claim an alternative
status to "homeless" more successfully, in large part because they have
recourse to the "battered woman" identity. Such an identity relies partially
on a feminist narrative that constructs the "battered woman" as a person
"in transition," a "survivor," stronger than other women for having experienced
and left a violent situation, or a participant in a "sisterhood" that includes
many other women assumed to have similar histories and needs.

Like other residents of Rose's House, Latanya, a twenty-eight-year-old
African American woman with two children, also does not consider herself
homeless, asserting that homeless people have different problems than she
does, particularly that they have "given up." Claiming that "there is always
a way out of homelessness," Latanya can never picture herself "on the street,"
and states that people become homeless because, unlike her, some are lazy,
others "like to live like that," and still others "have been hurt badly
in life." She argues that she would do anything to keep from becoming homeless,
such as collecting cans for money or relying on public welfare payments
until she could find work.

Despite her wish to distinguish herself from homeless women, Latanya's
own story supports the contention that many women seek shelter for a multiplicity
of reasons. Latanya grew up in a family with eight children, with a father
who was an alcoholic and abusive to her mother and to the children. Like
Latanya, most of her sisters have also had violent relationships, and she
remains close to only one sister who also lives in Phoenix. Latanya had
her first child at age seventeen, dropped out of high school, and lived
with her son's father for five years. When she left him, Latanya moved
with her son to an apartment subsidized by the Department of Housing and
Urban Development (HUD). With subsidized housing, work in a fast food restaurant
and cleaning houses, and some help from her sister, she managed to support
herself and her child.

Two years later, Latanya met her current boyfriend, Emil. After two years
together the couple had a son, and she gave up her subsidized housing a
year later to move into his home. Over the course of the past year, Latanya
left Emil several times and stayed with her sister when he became violent
toward her. Finally, she decided to leave him for good and called the shelter:
"My sister wanted me to stay with her, but I wanted counseling, so I went
to Rose's House." Although she claims that she is not "really" homeless
because she has the option of living with her sister, Latanya's sister's
husband already works three jobs to support the family, and two or three
more people would certainly strain their resources. Moreover, Latanya has
applied for subsidized housing again, but she faces a significant delay
in obtaining such housing because of a two-year waiting list.

Like Latanya, Betsy describes multiple, interlocking factors that led to
her residence first in a domestic violence shelter and then in a homeless
shelter. And like Latanya, Betsy does not clearly fall into either the
"battered woman" or "homeless woman" categories. A thirty-one-year-old
Euro-American woman with three children, Betsy begins the story of how
she became homeless when she ran away from home at age fifteen, saying,
"I guess I've been homeless from fifteen to twenty-one, but I didn't think
of myself that way then." For those six years, Betsy supported herself
through sex work and as a relief driver for truckers, riding back and forth
across the country with truckers who paid her one cent a mile to drive
while they slept. She had to sleep on the streets only twice during those
years, but supporting herself through sex work was difficult for her emotionally.
Her voice lowers to a whisper and she cries when she talks about it.

For a few years in her early twenties, Betsy worked alternately as a live-in
housekeeper and as a cashier in a retail store. She then met her husband
and had her first child, who is now four years old. Because the couple
had difficulties getting by on the money her husband made at his appliance
repair business, they were repeatedly evicted when they were short on rent.
In the four years prior to Betsy's arrival at the shelter, they had moved
twenty-two times. During the past year, cocaine had played an increasing
role in their marriage, and Betsy's husband had become increasingly violent.
As the couple became more involved in using drugs, his violence escalated
and he grew less interested in and committed to working. Betsy blames him
for their homelessness, arguing, "The main reason I'm homeless is that
Ron didn't want to work." Because Ron did not want Betsy to work, and three
births in the past four years kept her at home for some months with each
baby, the family's financial problems intensified. The last time they were
evicted, the landlord would not return what was left of their security
deposit for fourteen days, and they had no money to pay for a motel or
another apartment.

Betsy had tried to leave her husband some months earlier, after she felt
his drug use and the violence had gotten "out of control." However, she
asserts, "the only place I had to go was my father's house. I was trying
to quit using drugs, but he's an addict too, and they would knock on my
bedroom door at night saying, `Betsy, do you want to get high with us?'
I knew if I didn't get out of there I wouldn't be able to quit using."
With three children in tow, her friends were not willing to take her in
or able to spare the room or the funds to feed and house four more people
for any period of time, so Betsy returned to her husband. She tried for
the next two weeks to get into Rose's House, but either the shelter was
full or the staff did not think Betsy's "problem was domestic violence,
because I wasn't defining it that way. I didn't understand that was the
issue." Eventually, Betsy went to another organization that assists homeless
people, sometimes directing them to area shelters. They helped Betsy get
into Rose's House the following day by emphasizing her battering relationship.
She stayed there for the next six weeks, then went to The Family Shelter.


It is difficult to point to the one reason Betsy became homeless. Even
before she and Ron began using cocaine, neither had education nor skills
that translated to jobs that paid enough to adequately support their children
and themselves. The couple faced great difficulty finding low-income housing.
Their repeated evictions meant yet another obstacle to stable employment,
as they were constantly in the process of moving from one apartment to
another, sometimes a fair distance from their former residence and jobs.
Nor did they own a car to facilitate travel to work. Moreover, Ron tried
to start his own repair business, working out of his apartment, but continual
moves made it difficult for customers to locate him. Both had cultural
expectations that a husband would take care of "his" family, expectations
that made Betsy reluctant to work full-time and Ron less likely to encourage
her to do so. Thus, they had to forego a badly needed second income. On
the other hand, with three children, the cost of child care during work
hours would have demanded a large portion of their second income. Ron's
violence created even less stability in their household, and Betsy's inability
to rely on a family she had fled at age fifteen created few options for
her when she attempted to leave him.

Domestic Violence

Rather than approaching domestic violence as a direct (or the single) "cause"
of women's homelessness, this study considers how such violence can help
to create the circumstances that might particularly make a low-income woman
more susceptible to homelessness.21 Domestic violence and poverty may intersect
with other issues to produce the circumstances that often lead to homelessness,
creating conditions that leave women no other choice but to seek temporary
shelter for the short-term, and thereafter to remain precariously housed.
Because of the extensive resources a woman who leaves a violent relationship
needs in order to support herself and her children, the shelter presents
one of the few choices available to low-income women. It demands a significant
amount of money to avail oneself of other options, such as staying in a
hotel, renting an apartment alone, or moving to another city.

Ella's story provides an expression of the ways that domestic violence
and poverty may intersect to create the very elements that often lead to
homelessness. A forty-three-year-old Euro-American woman with three children,
Ella left her husband five years ago and entered Rose's House domestic
violence shelter. By the time she left, the violence in their eighteen-year
marriage had escalated to the point that her husband, Jim, often threatened
her with knives and guns. She believes that if she had stayed longer, he
eventually would have killed her, and as a result Jim still does not know
where she lives. During their marriage, the couple was homeless off and
on, at one time sleeping for four months in the woods and for two years
in their camper, getting most of their food from dumpsters. Ella worked
sporadically as a waitress and her husband in construction, but primarily
they lived isolated lives in rural areas. Ella notes that the poverty and
violence escalated simultaneously over time, to the point that in the last
year and a half of their marriage they rarely bought groceries: "One time,
we lived off canisters of candy thrown away by the supermarket for one
month. We lived off chicken thrown away from the supermarket deli. I would
have to pick off the dirt and mop strings before serving it." After she
left Jim, Ella stayed at Rose's House for two months.

The violence in Ella's marriage made a fragile economic situation worse
and contributed to her need for social services or a shelter when she left.
She needed a supportive shelter or program in order to ensure her and her
children's safety and allow her time to finish school and find employment.
Women like Ella who leave violent partners are much like other homeless
women who do not have family on whom to depend or whose families are too
impoverished to offer them many resources. Because an abuser often isolates
his partner from friends and family, and a woman's shame for the beatings
encourages her to remain isolated, women fleeing violence may lack a network
of friends and family upon whom to rely for housing, money, food, or clothing.
Moreover, low-income women who wish to leave such a relationship often
do not have any money saved that they can use to support themselves until
they find employment. Others cannot find a job that pays a living wage.
Even a woman leaving a middle-class home may have earned a salary well
below her husband's and may find it difficult to live on her income alone.
Although Ella possessed a high school diploma, during her marriage she
"felt like there was no way out . . . that it was possible only to get
waitressing and fast food jobs" that she did not think would support herself
and her children. At the same time, however, she began to realize that
she could obtain assistance from the social service organizations (from
which she and her husband had occasionally obtained food boxes) without
her husband, and she saw that as providing an escape from the marriage.


Ella currently juggles a full-time university schedule (she has a 4.0 average),
part-time employment as a case manager working with elderly men and women
as a counselor and advocate, and raising her three children, ages nine,
ten, and thirteen. In addition to subsidized housing, she receives a Pell
Grant and some support from AFDC as her work at the hospital is an unpaid
internship, a requirement for her graduation. She also participates in
a transitional program for homeless families called Endowment for Phoenix
Families (EPF). The program's goal is to help people become financially
secure enough to permanently leave behind government-subsidized housing.
In the three-year program, families find their own housing and pay a portion
of their incomes to the landlord for rent, while HUD pays the balance with
a voucher. Despite all her accomplishments, Ella and her children continue
to live in constrained economic conditions. Their old, sparsely furnished
house has torn window screens, and the carpet is worn through to the concrete
floor in several places. Ella explains, "we still run out of almost all
food except beans and rice by the end of the month."

In pointing out the influence of poverty on the lives and choices of women
currently living in domestic violence shelters, my interest is not to argue
that low-income men are necessarily more abusive than their well-off counterparts.
Rather, I am exploring the reasons that primarily low-income women go to
the shelters. This study is done in a context, however, of an association
of domestic violence with low-income people, which includes a sociological
approach to the analysis of battering that concentrates on various "subcultures."22
This is driven by the perspective that social variables, such as class,
race, ethnicity, gender, and age, influence the tendency toward violence.
As a result, it is argued that African Americans, Latinos, and members
of the working class have a greater tendency toward violence in their relationships
because of the "social stress" of higher unemployment rates and impoverishment.


This view, however, is not universally accepted. For example, Linda Gordon's
work suggests that both historical and current correlations between differing
rates of violence and specific groups of people may warrant further attention.
Gordon criticizes the historical view that "family violence is a problem
of poor people,"23 an understanding of violence in the 1930s that led organizations
to address poverty as a key to eradicating violence. Gordon cautions that
measuring "the contribution of economic stress to family violence"24 historically
has produced ambiguous findings. She comments on the inconsistency of 1930s
and 1940s statistics that correlate poverty with family violence, concluding
that in some cases "what is being measured is as much the sensibility of
caseworkers as the conditions of clients."25 As this paper argues, the
association of battering with racial "others" and impoverished people continues
to define popular cultural understandings of battering in the present.
This study supports Richard Gelles's suggestion that a critical perspective
like Gordon's might be brought to bear on discussions of battering today.
Gelles proposes that the higher incidence of violence found in present-day
Latino and African American families may be in part attributable to overreporting
relative to reporting on violence in white families.26

Although Gelles also maintains that current studies indicate the risk of
violence is higher for people under stress from unemployment and poverty,
his and Gordon's arguments should caution researchers of battering to consider
the complexities involved in creating measures and interpreting statistics
in this area. First, in terms of this study, a predominately low-income
population at La Casa and Rose's House may be traceable to the limited
options available to low-income battered women relative to higher-income
women wishing to leave their relationships. It does not necessarily suggest
a higher incidence of battering among poor people. Moreover, just as is
the case with many researchers and academics, caseworkers at domestic violence
shelters are overwhelmingly middle class and Euro-American. Their participation
in constructing meanings for domestic violence is influenced by their own
social locations and biases, in particular, their assumptions that abusive
relationships are most likely to occur in the families of racial "others"
and low-income people. This paper argues, then, that surveys measuring
the incidence of violence within certain groups of people are wrapped up
in cultural understandings of battering that depend to some extent upon
racialized and class-based assumptions about what "kind" of men beat their
wives or partners.

The existence of such assumptions invites further examination of the cultural
meanings of "abuser" and "battered woman." In addition to enumerating the
everyday experiences of the women interviewed for this study, it is important
to understand the ways in which reality is socially and culturally constructed
through narratives of domestic violence. Culturally available narratives
provide a way for people to understand battered women's experiences, a
shorthand that explains a woman's behavior and delineates her status as
"domestic violence victim" or "survivor," as "battered woman" or "not battered
woman." The subject positions available to a woman experiencing violence
are bounded by a limited number of narratives that tell us what domestic
violence entails and who is a "battered woman."

Much of the writing and public discourse on domestic violence suggests
that a woman who escapes a violent relationship is almost always in serious
physical danger and therefore needs a battered women's shelter in which
to hide from a mate who probably will stalk her and possibly eventually
kill her if she leaves him without going into hiding.27 In this narrative,
a woman remains in a battering relationship for fear of incurring increased
violence should she leave, and sometimes more importantly because, despite
her fear of her partner, she still loves him. Because he has decreased
her self-esteem28 to the point that the battered woman on some level believes
she does not deserve better treatment, she needs counseling. Without counseling
and attention to her emotional condition, in particular that of "learned
helplessness"" (in which a battered woman becomes increasingly depressed,
anxious, in denial about the seriousness of the violence, and less likely
to fight back), she is more likely to return to the relationship.

This narrative does not provide a space for a "battered woman" to be anything
other than "agent" or "victim," identities that are set up as binary opposites
with little middle ground or possibility that a woman may be both active
in her own defense and in pursuing strategies to end the violence and abused
mentally, emotionally, or physically.30 An either/or dichotomy between
helpless victimization and powerful agency equates agency with leaving
the relationship and psychological victimization and impaired self-esteem
with staying. Women involved in violent relationships may, however, be
active agents freeing themselves from abuse, many remaining in a relationship
while incorporating a "vast array of personalized strategies and sources
. . . to end the violence in their lives."31 Such strategies may entail
remaining in the relationship for a period of time while trying to negotiate
an end to the abuse, or to save money, or to finish school or job training
in order to leave with more resources. Requesting family networks to pressure
the abuser to end his violence, or turning to social service agencies for
assistance are other strategies.32

A competing feminist narrative argues that violence against women is part
of a cultural milieu that accepts and rewards male violence.33 According
to Kersti Yllo, a feminist perspective on domestic violence situates male
violence within a discussion of gender and power, thereby rejecting the
"gender neutral" sociological explanation of "family violence."34 Yllo
points out that the sociological perspective makes little of the fact that
relationship violence is primarily a "male phenomenon," that in cases where
women have reported violence toward their partners it generally has been
in self-defense, and that domestic violence is a "tactic of entitlement
and power that is deeply gendered" rather than about a "conflict of interest."35
Women protesting against male violence or trying to leave their relationships
may face indifference or disapprobation from the police, courts, the church,
and their families. As such, leaving a violent partner is complicated by
women's lack of community support and the possibility of increasingly violent
repercussions from abusers. Feminists and formerly battered women argue
that shelters were founded in response to women's need for protection from
their partners and for counseling and support from other women. A feminist
perspective argues that all women who have abusive husbands or lovers are
victimized, yet even women who return to their relationships may be "agents,"
as such women may be actively involved in trying to end their partner's
violence.

Women of color have argued that such alternative narratives on issues of
rape and battering offered by white feminists often ignore the specific
experiences of women of color, or worse, utilize the racist stereotyping
relied upon in the dominant narrative.36 In her study of battered women's
shelters in Los Angeles, Kimberle Crenshaw reveals the dynamics of "structural
intersectionality" as they play out in the lives of low-income African
American, Latina, and Asian American battered women, arguing that women
of color suffer multilayered subordination based on race, gender, and class.
Crenshaw suggests that persistent poverty and its attendant issues of housing
shortages and underemployment are particularly important in considering
the experiences of battered women of color:

Economic considerations-access to employment, housing, and wealthconfirm
that class structures play an important part in defining the experience
of women of color vis-a-vis battering.... shelter policies are often shaped
by an image that locates women's subordination primarily in the psychological
effects of male domination, and thus overlooks the socioeconomic factors
that often disempower women of color. Because the disempowerment of many
battered women of color is arguably less a function of what is in their
minds and more a reflection of the obstacles that exist in their lives,
these interventions are likely to reproduce rather than effectively challenge
their domination.37

Much like those women Crenshaw describes, the women who entered the domestic
violence shelters in Phoenix have few personal resources such as savings,
significant education and job training, or family or friends who can take
them in for any period of time. Most women have housing and job needs that
often are not addressed at the shelter. Rather, the shelter is designed
principally to shield women from danger and offer counseling and emotional
healing in hopes that a woman will not return to her relationship. African
American, Latina, and American Indian women interviewed in the Phoenix
shelters contend with racial discrimination in the past and present that
make their journeys to stable housing more difficult. Despite, then, the
construction of "battered woman" that intimates that all women leaving
a violent relationship need psychological counseling or are in serious
physical danger, many low-income women look to shelters to provide housing.
As Crenshaw argues, women of color are disproportionately low income and
face significant financial obstacles in leaving abusive relationships.


Staff members of Phoenix battered women's shelters present yet another
competing narrative of domestic violence, which shares some commonalties
with both the feminist and dominant narratives.38 In this interpretation,
like the feminist one, women are victims of controlling male violence and
need the shelter in order to hide from their husbands or lovers who are
likely to be more violent should they locate their escaped partners. Alternatively,
others need the shelter in order to work on their diminished self-esteem
while finding housing and a job. Shelter staff identify emotional, verbal,
and sexual abuse as domestic violence as well as physical abuse, and in
this sense recognize that all women seeking shelter do not necessarily
need protection from their abusers. Despite their claim that many women
do not need physical protection, however, some staff operate as if they
assume that most women do need the shelter for physical safety. For example,
some staff members argue that a woman who visits her abuser while she resides
in the shelter, thereby proving that she is not hiding from him and not
in severe physical danger, does not "really" need shelter. In other words,
she is not a "real" battered woman.

Although Rose's House and La Casa are not among them, two of the seven
domestic violence shelters in Phoenix have instituted "closed" programs,
based on the argument that both battered women and their partners are sometimes
"addicted" to their relationships. These shelters dictate that women entering
their programs remain literally on the shelter premises, eschewing any
contact with friends, employers, and especially their partners for a period
of several days to up to two weeks. Women must get a leave of absence from
work, reschedule any appointments or meetings during that time, and generally
upset their daily schedules. One woman called Rose's House wanting to transfer
there from a shelter with a closed program, complaining that she had already
missed one housing appointment. She had scheduled the appointment in anticipation
of leaving her husband and was in danger of missing more if she stayed
at the shelter. Thus, those women who have been working on separating from
their abusers even before calling a shelter, in particular by arranging
for housing, find themselves forced to put most practical considerations
aside in such closed programs.

Supporters of closed programs argue that a woman in a violent relationship,
much like a drug addict or alcoholic, is on some level addicted to the
relationship and needs a period of intense "in-patient" counseling in order
to ensure her "recovery." By mandating resident confinement for a period
of time, closed programs strive to "break the addiction" to an abusive
partner, suggesting that at the end of a period of intensive counseling,
a woman will be less likely to return to her partner. The rationale for
canceling appointments for low-income housing or demanding that a woman
miss work lies in the argument that a battered woman's most pressing need
is for counseling, supported by the notion that women remain in abusive
relationships primarily because of low self-esteem, codependence, or fear
of being alone. While these psychological factors may affect a woman's
decision about whether to leave her relationship, closed programs basically
ignore the economic insecurity that many Phoenix women factor into their
decisions about their relationships. Likewise, even in programs that are
not closed, women are sometimes not allowed to accept a job if it interferes
with regularly scheduled weekly or nightly group counseling meetings.

In addition, like the dominant narrative, shelter workers tend toward a
dichotomization of agency and victimization. Caseworkers at Rose's House,
for example, may suspect that those women who do not appear "helpless"
or have "low self-esteem" might simply need housing from the shelter, or
might be lying or not telling the whole story about their experiences with
abuse. In fact, I discovered that many women were primarily seeking housing
from the shelter. Although they, as well as other women interviewed, said
that the counseling and group support at the domestic violence shelter
helped them, they had entered the shelter because they had few options
regarding a place to live and financial support for themselves and their
children.

Marta's story, like Ella's, indicates the centrality of both domestic violence
and poverty to women's homelessness. Moreover, Marta describes the extent
to which economic dependence played a role in the dynamics of her battering
relationship. A forty-year-old Latina, Marta was staying with her five
children at The Family Shelter when I met her. She had tried leaving her
twenty-year marriage several times in the past, but, because she did not
know that shelters existed, Marta lived in her car with her youngest child
for five months while attempting to become financially independent of her
husband. Unable to make the transition from the car to housing, she returned
and lived with her husband for a few months. During a fight between Marta
and her husband that resulted in a fire in the house, the police were called,
and they provided Marta with a list of battered women's shelters. She applied
to Rose's House and was accepted within a week, staying for a month before
moving to The Family Shelter.

The day I met Marta, she had only two weeks left in the shelter before
her ninety days expired. When she spoke about her search for housing or
another program for her family, her face clouded with worry: "I have too
many children to get into the Endowment for Phoenix Families. Other people
have found me eligible, but no one has a place open." Nor could Marta find
an apartment building that would accept her, even though she had enough
money saved to pay her deposits. As soon as the managers learned that she
had five children, lived on an income of $561 per month from AFDC, and
had a problematic credit history, they refused to rent to her. The caseworker
at The Family Shelter was pressuring Marta to ask her husband for money
to help support their children. Indeed, her stay at the shelter for the
next two weeks was contingent on her demanding money from her husband.
Marta did not want to involve him in her financial difficulties, believing
that a request for his assistance would be the first step in returning
to her husband. In thinking about her options, Marta remarked, "I'm in
a position now where I will remain homeless or go back with my husband."
She did not want to live with him again, saying that she had spent a weekend
with him and saw his "old behaviors coming up" and felt that it was "dark
and tense" at his apartment.

In the ensuing weeks, shelter staff allowed Marta to stay beyond their
usual ninety-day limit, but she still could not find housing. She spoke
with increasing anxiety about her inability to find a place to live and
spent hours sitting by the phone in the community area near the office,
making calls and checking messages repeatedly to see if an apartment manager
or program director had returned her calls. When Marta's extensions at
the shelter had run out, she refused to return to her husband. A month
after the first interview she was sleeping in her car and her children
were sleeping in a tent trailer behind an acquaintance's home. As Marta's
story illustrates, the scarcity of low-cost housing plays a significant
role in narrowing the choices available to battered homeless women trying
to live on their own. Veronica's and Lisa's experiences have some parallels
to Marta's in that their inability to pay for housing and other living
costs without a male partner's assistance critically affected the choices
available to them. Indeed, a number of Phoenix women were forced to enter
shelters when they found that their low-wage jobs or AFDC benefits were
not enough to pay for rent, utilities, food, and clothing. Veronica is
a twenty-five-year-old Latina with two children, ages two and four, living
at La Casa:

I came to the shelter yesterday. I was living with my boyfriend on and
off for six years.... I left him before and stayed with my mom. I went
back because I didn't get along with my mom, and I didn't have anywhere
else to go. The last time he beat me up, I called the cops for the first
time-he's in jail now. I was tired of the abuse, and I felt like if he
didn't care about me, if he could do that to me, I shouldn't care about
him.

Veronica receives $347 per month from AFDC for her two children, with which
she is supposed to pay for rent, utilities, and other necessities.39 With
an eighth-grade education, she possesses few marketable skills. She says
she lied to the welfare department, telling them that she did not live
with her boyfriend in order to receive benefits. Even with benefits, however,
she and her boyfriend survived only because his father took them in and
accepts $150 per month for rent and utilities, which, according to Veronica,
"was nothing compared to getting your own place." She had attempted to
rent her own apartment in the past: "I tried to get a place on my own,
but I couldn't keep up with the electricity payments, so I went back to
my boyfriend." Another time, she had tried to leave her boyfriend and live
with her father, but his new wife did not want her and her children to
live with them. Veronica hopes that the shelter staff can help her find
housing and pursue her GED and computer training. She remarked sadly, "My
boyfriend used to threaten me that if I left him, he'd come find me....
Now he says he beats me because he wants me to leave."

Though she has more resources than Veronica, Lisa, a Euro-American woman
in her late thirties, experienced similar housing problems when she came
to Phoenix with her two children, ages eight and four, to hide from a fiance
she describes as "paranoid and abusive." When she realized that her fiance
would try to track her down if she left him, Lisa decided to move to a
city where she had no family or friends, so he would be less likely to
find her. She and her children arrived with no car, no employment lined
up, and little money, since in the last months of their relationship Lisa
had depended on her boyfriend financially while she tried to start her
own business. After two weeks in Phoenix during which Lisa stated she had
not had a full meal, she resigned herself to the need for assistance and
applied for AFDC: "People always say that getting on AFDC is a cop-out.
The easy way would have been to stay with [my fiance], but it took more
integrity to leave and get on assistance."

Unable to afford an apartment on her AFDC allowance, Lisa rented a room
for herself and her children in a house owned by a woman who lived with
her emotionally disturbed young son. However, the mother and son's constant
physical fighting caused Lisa and her children considerable stress. Moreover,
the woman began locking the phone and answering machine in her room, both
of which were crucial in Lisa's efforts to develop a career as a freelance
writer. The family then moved in with an older woman, who seemed to be
relatively stable but who Lisa soon learned was a "closet alcoholic." Complaining
that the family was home too much, after a month the woman was no longer
willing to rent to them.

By this time, Lisa had made a friend who owned his own business in Phoenix
and arranged for her and her children to sleep in his office until they
were able to find better housing. They slept on the floor in the unheated
office for the months of December and January, piling on blankets to try
to stay warm. Lisa finally decided to look into shelters, though she feared
a drug-infested and unsafe environment. She met with a case manager from
The Family Shelter, who, she asserts, had to convince her to come to the
shelter: "I thought a shelter would be so unsafe that I'd rather be on
the street than there. I agreed to go because of rules like no drugs or
alcohol [allowed on or off the shelter premises], people have to sign in
and out, and no one's allowed in the rooms." Surprised by the structure
and support offered at The Family Shelter, Lisa moved her children there
in early March. From there, a longer-term, transitional program for homeless
families accepted her. Analyzing her struggles over the past year, Lisa
asserts that women's ability to leave abusive partners "all comes down
to economics": "So many women stay in abusive situations for economic reasons....
I tried for a year to get on my feet, but you can't do it without a stable
living environment.... It was a blow to my pride and self-esteem to end
up in a program like this."

Describing her life before moving to Phoenix as a combination of generally
stable periods alternating with tumultuous intimate relationships, Lisa
has been married twice, both times to abusive men. After her children were
born and Lisa had divorced her first husband, she completed a college degree
while holding down a job. Though she had some difficulty supporting her
family during those years, since completing her degree five years ago Lisa
has managed to live a relatively comfortable lower-middle-class life. With
her college degree, she worked as a freelance desktop publisher, which
did not afford her a consistent or wholly reliable income. Still afraid
that her fiance will find her, since moving to Phoenix Lisa has contacted
only one family member, her sister, though even she does not know where
Lisa lives.

Like Lisa, Veronica, and Marta, many Phoenix women describe significant
obstacles to finding housing that is both affordable and safe. With more
than three low-income renters for every low-cost apartment available in
Phoenix, compounded by lengthy waiting lists for subsidized housing, the
housing shortage contributes enormously to women's vulnerability to homelessness.
A low-income woman's problems with securing an income that pays a living
wage, and thereby an apartment, may be complicated by her wish to avoid
her ex-husband or partner and in most cases by her primary or sole responsibility
for financial support of her children.

Conclusion

Women in domestic violence and homeless shelters tell notably similar stories
about their paths to emergency shelters, regardless of whether those shelters
are for battered women or homeless women. Listening to their stories suggests
that the distinctions drawn between the identities of "homeless woman"
and "battered woman" are often arbitrary. The organization of homeless
and domestic violence shelters, however, demands that women define themselves
in one way or another in order to receive services. Women who cannot find
an opening in a battered women's shelter may have to conceal their histories
with abusive partners in order to gain entrance into a homeless shelter.
Other women may repackage themselves in order to receive assistance from
a battered women's shelter.

Shelters reinforce the distinction between homeless and battered identities
through the services they offer. Notwithstanding the pattern of persistent
poverty and battering, homeless shelters are likely to focus on women's
housing and employment needs, while battered women's shelters concentrate
on the psychological ramifications of violence, often to the exclusion
of providing staff time to help with housing and job searches. In addition,
both kinds of shelters have three-month time limits that preclude serious
attention to women's skill levels and training, making receipt of subsidized
housing-with waiting lists of up to two years-an undependable option. Although
women usually receive federal or state preferences once they enter a shelter
and so will be placed at the top of most housing lists, three months may
easily pass before housing becomes available. Because battered women's
shelters define their services as "crisis" housing, most initially accept
women for no more than thirty days so women cannot always depend upon even
ninety days of shelter.

Women counter the story of homelessness suggested by such a shelter system
by emphasizing their persistent poverty and precarious housing arrangements
in concert with battering relationships. Women in both kinds of shelters
express their desires to receive respectful, individualized treatment from
a staff who will respond to their experiences with battering and their
housing and employment needs. However, when their ability to gain entrance
into a shelter depends upon their conformity to certain identities-"battered
woman" or "homeless woman"-women in Phoenix shelters will self-consciously
manipulate their histories in order to fit into these socially constructed
categories of need.

Notes

See Jan L. Hagen, "Gender and Homelessness," Social Work 32 (July/August
1987): 312-16; Joan Zorza, "Woman Battering: A Major Cause of Homelessness,"
Clearinghouse Review 25:4 (1991): 421-29; David Wagner, Checkerboard Square:
Culture and Resistance in a Homeless Community (Boulder: Westview Press,1993);
and Doug A. Timmer, D. Stanley Eitzen, and Kathryn D. Talley, Paths to
Homelessness: Extreme Poverty and the Urban Housing Crisis (Boulder, Westview
Press, 1994).

Linda Gordon, Heroes of Their Own Lives: The Politics and History of Family
Violence (New York: Viking, 1988); Kimberle Williams Crenshaw, "Mapping
the Margins: Intersectionality, Identity Politics, and Violence Against
Women of Color," in The Public Nature of Private Violence: The Discovery
of DomesticAbuse, ed. Martha Albertson Fineman and Roxanne Mykitiuk (New
York: Routledge, 1994), 93-118; and Ann Jones, Next Time, She'll Be Dead.
Battering and How to Stop It (Boston: Beacon Press, 1994).

3. Lee Ann Hoff, Battered Women As Survivors (New York: Routledge, 1990).
4. Hagen, "Gender and Homelessness."

5. U.S. Department of Commerce, Bureau of the Census, Income and Poverty:
1994

(Washington, D.C.: U.S. Government Printing Office, CD-ROM). 6. Laura DeKoven
Waxman, Kimberly Peterson, and Matthew McClure, A Status Report on Hunger
and Homelessness in America's Cities: 1995 (Washington, D.C.: The United
States Conference of Mayors, 1995), 70.

7. Laura DeKoven Waxman and Shannon Hinderliter, A Status Report on Hunger
and Homelessness in America's Cities: 1996 (Washington, D.C.: The United
States Conference of Mayors, 1996), appendix.

8. For example, Mark Robert Rank, Living On the Edge: The Realities of
Welfare in America (New York: Columbia University Press, 1994).

9. National Coalition for the Homeless, A Place Called Hopelessness: Shelter
Demand in

the 90s (Washington, D.C.: National Coalition for the Homeless, n.d.),
21. 10. The names and identifying background information of interview participants
(both shelter staff and residents) and that of homeless and domestic violence
shelters have been changed in order to protect the anonymity of participants.
11. Aihwa Ong, Spirits of Resistance and Capitalist Discipline: Factory
Women in Malaysia (Albany: State University of New York Press, 1987). 12.
Shelters that receive funding from the Department of Economic Security
may pro

vide ninety days of shelter per person during each fiscal year. 13. For
a more detailed description of the environments and rules of the shelters,
see Jean Calterone Williams, "Geography of the Homeless Shelter: Staff
Surveillance and Resident Resistance," Urban Anthropology 25:1 (1996).


14. For example, Peter H. Rossi, Down and Out in America: The Origins of
Homelessness (Chicago: University of Chicago Press, 1989); Robert C. Coates,
A Street Is Not a Home: SolvingAmericas Homeless Dilemma (Buffalo, N. Y.:
Prometheus Books, 1990); and Christopher Jencks, The Homeless (Cambridge:
Harvard University Press, 1994). 15. Peter Marcuse, "Neutralizing Homelessness,"
Socialist Review 18 (1988): 88. 16. Wagner, Checkerboard Square, 5.

17. For a discussion of the *experiences and characteristics" associated
with battered women, see Donileen R. Loseke, The Battered Woman and Shelters:
The Social Construction of Wife Abuse (Albany: State University of New
York Press, 1992). 18. Jones, Next Time, She'll Be Dead, 83.

19. Florence Hollis, Women in Marital Conflict: A Casework Study (New York:
Family

Service Association of America, 1949).

20. Lenore Walker, The Battered Woman (New York: Harper and Row, 1979),
"The Battered Woman Syndrome Is a Psychological Consequence of Abuse,"
in Current Controversies on Family Violence, ed. Richard J. Gelles and
Donileen Loseke (Newbury Park, Calif.: Sage Publications, 1993), 133-53.
21. Timmer, Eitzen, and Talley, Paths to Homelessness.

22. Richard J. Gelles, "Through a Sociological Lens: Social Structure and
Family Violence," in Gelles and Loseke, Current Controversies on Family
Violence, 31-46. 23. Gordon, Heroes of Their Own Lives, 151. 24. Gordon,
Heroes of Their Own Lives, 149. 25. Gordon, Heroes of Their Own Lives,
149. 26. Gelles, "Through a Sociological Lens." 27. Loseke, The Battered
Woman and Shelters.

28. Walker, The Battered Woman; and Del Martin, Battered Wives (New York:
Simon

and Schuster, 1983 ).

29. Walker, "The Battered Woman Syndrome."

30. Martha R. Mahoney, "Victimization or Oppression? Women's Lives, Violence,
and Agency," in Fineman and Mykitiuk, The Public Nature of Private Violence,
60-63. 31. Lee Bowker, "A Battered Woman's Problems Are Social, Not Psychological,"
in Gelles and Loseke, Current Controversies on Family Violence, 155. Also
see Gordon, Heroes of Their Own Lives; and Mahoney, "Victimization or Oppression?"
62. 32. Susan Schechter, Women and Male Violence: The Visions and Struggles
of the Battered

Women's Movement (Boston: South End Press, 1982), 232-33. 33. Kersti Yllo,
"Through a Feminist Lens: Gender, Power, and Violence," in Gelles and Loseke,
Current Controversies on Family Violence, 47-62; Gordon, Heroes of Their
Own Lives; and Schechter, Women and Male Violence. 34. Yllo, "Through a
Feminist Lens."

35. Yllo, "Through a Feminist Lens," 57. See also Gordon, Heroes of Their
Own Lives. 36. Kimberle Crenshaw, "Whose Story Is It Anyway? Feminist and
Antiracist Appropriations of Anita Hill," in Race-ing Justice, En-gendering
Power: Essays on Anita Hill, Clarence Thomas, and the Construction of Social
Reality, ed. Toni Morrison (New York: Pantheon Books, 1992), 402-40.

37. Crenshaw, "Mapping the Margins," 95-96.

38. Schechter has written of the tensions involved in defining the components
of battering, tensions that increased as shelters and shelter staff professionalized
in the late 1970s and early 1980s. In Phoenix shelters today, these tensions
sometimes play out as a battle between feminist understandings of battering
and understandings based on the model used by professional social workers.
Social workers became the primary staff at shelters as public and private
grants to run shelters became more available. See Schechter, Women and
Male Violence, 101-3.

39. The poverty level, as calculated by the Department of Health and Human
Services in 1994 was $ 12,320 per year for a three-person family (see Federal
Register vol. 59, February 10,1994, 6277-8). However, the maximum median
state AFDC grant for the same size family was $4,392, excluding food stamps
and other in-kind benefits such as Medicaid.