The Homeless in America: An Examination of Possessions and Consumption Behaviors

This research utilized an ethnographic approach to advance our understanding of the survival strategies
employed by the homeless in our society. We examine the types of possessions consumed, how
possessions are acquired through nontraditional employment and scavenging, and why some
products are purchased while others are scavenged. We also look at the tools used to facilitate search,
acquisition, storage, and consumption of these products. Finally, we consider the importance of
community for protection of self and possessions and how community among the homeless affects
consumption. Emergent themes that allow interpretation of the description are presented.

During the 1980s, homeless men, women, and children began crowding urban America's back
alleys and streets (King et al. 1989, National Mental Health Association 1988). Current debates
center around the makeup of the homeless population and its size. Government-sponsored
research suggests that the homeless are made up of deinstitutionalized mental patients, drug and
alcohol abusers, families with a black or Hispanic woman as head of household, and the
marginally employed who have suffered a major financial setback, such as with a prolonged illness
(U.S. Department of Health and Human Services 1984). However, few sources agree on the
precise representation of these groups among the homeless. For example, Torrey (1988), author
of Nowhere to Go, suggests that recent increases in homelessness are due primarily to a public
policy of deinstitutionalization of the mentally ill, who are released into communities unprepared
to provide adequate support. Conversely, Wright (1988) reports that the true rate of mental ill-
ness among the homeless is only 10-33 percent (see also Snow et al. 1986).

There is also controversy surrounding the actual number of homeless persons (Ropers 1988).
Democratic leaders, such as former presidential candidate Michael Dukakis and former Speaker of
the House Jim Wright, have claimed that 3 million Americans are huddled in the streets, doorways,
and shelters each night (Whitman 1989). However, a 1988 study by the nonpartisan Urban Institute
concluded that, at most, 600,000 Americans were homeless on any given day in 1987.

While these concerns are important, another set of pertinent issues involves the struggle by the
homeless for survival (Hirsch and Stamey 1988). To date, no attempt has been made by
consumer researchers to investigate what possessions are deemed necessary for survival, how the
homeless acquire these possessions, and whether these possessions are consumed in isolation or
within some form of community. The purpose of this research is to provide preliminary answers to
these questions.

First, we turn our attention to the plight of the homeless and provide a definition of homelessness
that establishes a framework for understanding their unique consumer-behavior situation. Second, we
describe our research method, which is based on the ethnographic tradition used in many studies
involving the homeless (Koegel 1987). Third, we present our findings by describing the types of
possessions consumed as well as how they are acquired and consumed by homeless per- sons.
Finally, emergent themes are provided to aid in the interpretation of this research, and consumer-
behavior implications are summarized.

THE HOMELESS

Prevailing Causes of Homelessness

Unemployment. Although the rate of unemployment in the United States is at a relatively low
level (approximately 5.4 percent), joblessness among the poor and minority groups remains at a
higher percentage (see Freeman and Hall 1987). Further, some states underestimate the
joblessness rate by reporting as "employed" anyone who works one or more hours a week or those
who have exhausted their unemployment benefits (Ropers 1988). Statistics obtained from
municipal shelters in New York City show that 40 per- cent of current occupants report loss of a job
as the cause of their seeking shelter (Salerno, Hopper, and Baxter 1984). Part of this problem may
be a result of the loss of millions of well-paying, unionized, semi- skilled manufacturing jobs and
their replacement by low-level service jobs. Hirsch and Stamey (1988, p. 5) state that "the unionized
manufacturing jobs which have been lost provided unique opportunities for high wages to those
with lower levels of skill and education. Newer low-level service jobs-such as hospital order- lies,
typists and word-processors, retail check-out clerks, waiters and waitresses, messengers for financial
firms, chambermaids in hotels-are not unionized and often pay below poverty level wages even for
full-time, year-round workers." Thus, the incentive to pursue employment by those at the lower
economic and social strata in society has been considerably reduced in re- cent years.

Deinstitutionalization. The deinstitutionalization movement of the past three decades resulted in

the discharge of hundreds of thousands of former patients of state mental institutions into unwilling
and unready communities (French 1987). Between 1955 and 1985, the number of psychiatric
patients in hospitals nationwide dropped from 559,000 to 138,000 (Greer 1986). Originally, it was
expected that treatment would be transferred to outpatient clinics. However, fewer than 800 of the
2,000 community mental health centers estimated as necessary to service this population have
been built, and the existing ones are not adequately coordinated with those institutions discharging
patients. According to Hombs and Snyder (1983), 1,000 mental patients are released from
hospitals nationally each day, and only 7 percent are referred to these centers. As a result, many
patients end up on the streets with no assistance and no alternatives.

Drug Addiction. Substance abuse, especially alcoholism, remains a persistent cause of homelessness.'
Recent investigations show that 33-38 percent of homeless adults are alcoholics, and 13-25
percent are drug abusers (Whitman 1989). Further, these dependencies tend to exacerbate typical
adverse health con- sequences of homelessness, such as nutritional and gastrointestinal disorders
(Schutt and Garrett 1988). For example, individuals with addictive disorders will often spend what
little money they have on such sub- stances rather than on food, clothing, or shelter.

Further, addictive problems may be one of the primary causes of homelessness among families
(see Fabricant 1988; French 1987), and drug and alcohol abuse may result in physical violence,
including spouse battering and sexual abuse (Hagen 1987a; Ropers 1988). Frequently, such violent
behavior by an adult male will cause a woman to take her child(ren) and leave, which renders
them homeless. Because most of these women come from poverty, they are unable to turn to
their extended families for financial assistance. Occasionally, both parents have addictive
disorders, which may prompt children to seek alternative living arrangements on the streets
(Hagen 1987b).

Scarcity of Low-Cost Housing. According to recent estimates by the National Housing Law Project,
the number of people who have been involuntarily re- moved from their homes is 2.5 million
(Salerno et al. 1984). Most of this displacement is the result of city "revitalization" projects and
economic development schemes (King et al. 1989) that allow occupied buildings to deteriorate to the
point of being uninhabitable before renovation for use by those with higher incomes. At the same
time, half a million units of low-cost housing are lost each year through conversion, abandonment,
arson, and demolition (Ropers 1988). Hirsch and Stamey (1988, p. 6) report that in New York City
alone, "the combined impact of gentrification and abandonment was to displace 225,000 city
residents a year; most of these are low-income families since both gentrification and abandonment
disproportionately affect poor people. The people displaced can find nowhere to go to live in part
because the same process is destroying low-income housing units at the rate of nearly one
hundred thousand a year."

This trend probably will continue. Throughout the 1970s and well into the 1980s, housing costs
rose at a rate greater than inflation (Hartman 1983; Rlossi and Wright 1987). Unfortunately, this rate
is more than twice the percentage increase in household income among renters during the same
time period (Dolbeare 1983; Ropers 1988). David Schwartz and John Glascock of the American
Affordable Housing Institute at Rutgers University estimate that an additional 4-14 million
American families are "now living on the knife edge of homelessness; they are doubled and tripled up
in the (mostly overcrowded and deteriorating) apartments of friends and family; they are one
paycheck, one domestic argument from the streets" (Rich 1989, p. A19).

The Homeless as Consumers

Homelessness has been defined as a lack of shelter that meets minimal health and safety standards
(Bachrach 1984; Caro 1981), and the definition includes those living squatter style in vacant
housing, stores, cars, vans and buses, and makeshift structures, or living on the streets (Nassau-
Suffolk Coalition for the Homeless 1989). However, this view is too narrow for consumer-behavior
purposes in that it does not encompass the full range of needs that are difficult for the
homeless to meet. Other researchers have suggested that poverty among the homeless inhibits
their ability to acquire not only adequate shelter but also food, clothing, medical care, and a host
of other goods and services necessary for physical and mental health (Freeman and Hall 1987;
Hirsch and Stamey 1988).

With regard to such products, Belk (1988, p. 139) states, "We cannot hope to understand consumer
behavior without first gaining some understanding of the meanings that consumers attach to
possessions." For the homeless, possessions and consumption behaviors that ordinary consumers take
for granted are often unavailable, reduced, or restricted. According to Koegel, Farr, and Burman (1986,
p. 133), "finding themselves homeless, their energies become focused on survival-on finding a
place to sleep and getting food into their stomachs. These tasks become full-time endeavors in and
of themselves; they have no time to do any- thing other than seek those things." Thus, our research
objectives were to (1) investigate the unique consumer- behavior activities of the homeless in terms
of what possessions are acquired as well as how they are acquired and consumed and (2)
develop a deep under- standing of the meanings of these possessions and behaviors and their
importance to the sense of self of homeless persons in our society.

METHOD

The use of the ethnographic research method has received some degree of acceptance in
consumer-behavior research recently (see Belk, Sherry, and Wallendorf 1988; Belk, Wallendorf,
and Sherry 1989; Hudson and Ozanne 1988). However, ethnography has been utilized for many
years, particularly by cultural anthropologists and, to a lesser extent, by sociologists (Berg 1989;
Fetterman 1989). Despite differences in terminology, most social scientists agree that the practice
of ethnography places researchers in the midst of whatever it is they study. Thus, Wolcott (1973)
has defined ethnography as the science of "cultural description," Geertz (1973) suggests that the
researcher's task is to provide "thick description," and Ellen (1984) characterizes the ethnographic
process as "subjective soaking."

Important criticisms have been raised concerning the ability of such naturalistic inquiry to produce
scientific knowledge consistent with the criteria posed by the positivist approach (Calder and Tybout
1989; Hunt 1989). According to Wallendorf and Belk (1989, p. 70), the positivist criteria of internal
and external validity, reliability, and objectivity can be substituted with the following
characteristics, originally formulated by Lincoln and Guba (1985), to maintain scientific integrity:
(1) credibility (adequate and believable representations of the constructions of reality studied); (2)
transferability (the extent to which working hypotheses can be employed in other contexts on the
basis of an assessment of similarity between two contexts); (3) dependability (the extent to
which interpretation is constructed in a way that avoids instability other than the inherent
instability of a social phenomenon); and (4) confirmability (the ability to trace a researcher's
construction of an interpretation by following the records kept). It is also clearly necessary that
the interpretation be unimpaired by lies, evasions, misinformation, or misrepresentations by
informants.2

Lincoln and Guba (1985) further suggest the following research techniques to improve the ability
of ethnographic research to meet criteria of integrity: prolonged engagement and persistent
observation; triangulation of sources, sites, methods, and researchers; regular, on-site team
interaction; negative case analysis; debriefing by peers; member checks; seeking limiting exceptions;
purposive sampling; reflexive journals; and independent audit. A discussion of the extent to which
each of these techniques was utilized during this investigation follows.

Prolonged Engagement Prolonged Observation and Persistent

Prolonged engagement and persistent observation are recommended to understand the broader
culture in which the phenomenon under investigation is imbedded and to provide the researcher
with the depth of knowledge necessary to recognize potential distortions in perception during the
course of data collection. Wallendorf and Belk (1989) recommend that researchers employ
perspectives of action-self-reports of behavior to the investigator by informants-as well as
perspectives in action-actual observations of in- formants' behaviors in their natural environment.
(This technique was originally developed by Gould et al. [1974].) Finally, they recommend honest and
open discussion with informants regarding the research purpose to allow access to a wide range
of relevant consumer behaviors.

Utilizing this technique, the researchers spent more than 1,000 hours in the field, developing a
deep understanding of the way of life of the homeless. Field- work began in early 1985 and
continued without interruption until the middle of 1989. During the early months of this project,
the focus was on learning how to build rapport with homeless individuals, developing an approach
to data collection on the basis of this rap- port, and discovering the nuances of the survival strategies
of the homeless.

These tasks were not easily accomplished. Many homeless persons are fearful of strangers because
they suffer such abuses as assault (e.g., "bum burnings" by teenagers) and rape and they fear that
their possessions may be stolen or confiscated. Thus, the mannerisms, dress, language, and behavior
of the researchers were tailored to reduce fear levels. Informal clothing and language, slow
movements, and an open and honest description of who the researchers were and why they were
"visiting" were employed. Often, providing a few dollars, something warm to eat or wear, or
sharing a bottle (making sure never to wipe the opening before drinking) would relax the informant
and produce more detailed descriptions or allow access to possessions and communal behaviors.
Further, the research team included both males and females to reduce the sense of possible
threat, particularly among homeless women, and visits to the same sites at regular intervals allowed
trust to develop between the researchers and info

Certain ethical issues also guided these interactions. As we became keenly aware of the "hidden"
homeless, we were careful not to divulge their whereabouts to others, especially the police. Thus,
we took every pre- caution to make sure that we were never followed on return visits. Further, our
field notes contained full descriptions of these encounters but avoided any at- tempt to place a
particular individual (at least by full name) in any specific location. Finally, we approached our
informants with compassion and understanding, avoiding any judgments that might further erode their
self-esteem. mants.

Triangulation of Sources, Sites, Methods, and Researchers

According to Wallendorf and Belk (1989), triangulation across sources and sites requires that the
re- searcher collect data from several informants in the variety of settings where the behavior
under investigation regularly takes place.3 Triangulation of methods dictates the use of multiple
forms of primary data col- lection, including field notes, still photography, tape recording, and video
recording. Finally, triangulation across researchers demands the use of a team of re- searchers
who provide interpretations of interactions with informants on the basis of their own frames of
reference or training.

To meet these requirements, several procedural steps were employed. First, the researchers collected
data from more than 100 informants in a variety of settings, including abandoned buildings, bridge
abutments and tunnels, shantytowns (i.e., small communities of makeshift dwellings), public parks,
and automobiles used primarily for shelter in both urban and suburban areas.

The urban locale in this research is a large north- eastern city where homelessness is considered
a growing problem due to an influx of illegal aliens and the unemployed from other parts of the
nation, as well as because of persistent poverty in several areas within the city. The sites selected
by our research team were the regions dominated by the poor and destitute, and these usually
were characterized by inadequately maintained buildings and almost nonexistent public services.
Crime, especially stealing, arson, and drug use, is a continuous problem for all residents of these
areas, including the homeless.

The suburban environment used is a middle-class area located in the center of a northeastern state.
This suburb is characterized by moderate crime but relatively high unemployment due to the
decreased emphasis on the manufacturing sector of our economy. Homelessness in this
community is considered a growing problem and exists on the fringes of the populated areas, in
public parks and forested areas and in the older, more dilapidated sections of the community.
Second, data were collected with hundreds of audiotapes and thousands of photographs, as well
as videotapes on several occasions. Field notes were maintained to record information not easily
observable with the other methods of data gathering. Third, the two primary researchers, each
trained in different disciplines (marketing/consumer behavior and sociology), provided separate
interpretations of the data to achieve triangulation across researchers.

Regular, On-Site Team Interaction and Negative Case Analysis

Regular, on-site team interaction requires that the researchers meet routinely to develop a
greater understanding of the range of possible interpretations of behaviors observed and
information obtained in the field. Wallendorf and Belk (1989) suggest that such meetings take
place after each member of the team has had an opportunity to generate personal interpretations
of the events to be discussed. One goal of such interaction is to produce a so-called negative case. A
negative case would involve interactions with informants that do not support the current set of
perspectives generated by the researchers. Therefore, interactions with informants in the field should
include an active search for such scenarios.

To ensure regular interaction between the primary researchers, discussions were held after each
contact with informants. The investigators formed their own private interpretations of the events
before' these meetings through the continuous use of reflexive journals, and discussions
highlighted differences of opinion. Although formal negative cases were not developed, an
attempt was made to continue gathering information in the areas where differences of opinion
regarding experiences in the field existed.

Debriefing by Peers and Member Checks

Wallendorf and Belk ( 1989) suggest that the primary investigators employ researchers not directly
involved with the project to critique and question the developing interpretations of the behaviors
under scrutiny. This support group should contain representatives from as wide a variety of
disciplines and backgrounds as appropriate to produce a diversity of opinions. Member checks
require a similar critique from a group of in- formants in the field. Precautions should also be taken to
include a variety of individuals as heterogeneous as the environment under investigation.

Debriefing by peers was accomplished in this project by providing a group whose professional
training included sociology, psychology, social work, anthropology, and marketing (consumer
behavior) with various perceptions of the field as well as materials used in the formation of these
perceptions over the course of the development of our final interpretation. Their reactions and
stated uncertainties provided an ongoing in- dependent audit of data-collection activities and were
used to guide additional data gathering.
Member checks were performed by providing homeless individuals with a description of how
others like themselves behaved (based on our data) and asking them for comments. Occasionally,
scenarios would be purposefully inaccurate to test the honesty of their re- actions. For the most part,
homeless individuals were quick to provide opinions regarding their peers and demonstrated
highly capable powers of observation that had been developed, in part, as a result of their need
to protect themselves and their possessions from harm and to remain alert for new sources of goods
and services.

Seeking Limiting Exceptions

This technique requires that the researchers progressively expand the number of sites and
groups included in their data collection to understand the limits of the transferability of their
findings. In this study, we found that our findings were limited to the homeless who live on the
streets rather than in public shelters. Most homeless persons differentiate themselves from shelter
inhabitants and feel that shelters are the choice of the most destitute or the insane. Thus, our
findings concern the independent-minded homeless who believe that they are surviving by their
own wits and re- sources. However, our informants varied widely demographically: 76 percent
were males and 24 percent females; 14 percent were under 30 years of age, 46 percent were 30-
40 years of age, 34 percent were 40- 50 years of age, and 6 percent were older than 50 years old; 42
percent were white and 58 percent black.

RESULTS

Our inquiry began with a general approach to data collection that investigated how and what
possessions are acquired as well as the environment in which these possessions are consumed by the
homeless. Also, we attempted to discover the effects of alternative consumer behaviors on the
meaning of possessions and their subsequent impact on the sense of self of the homeless.
Consistent with the interpretivist tradition, subcategories of these research directions were
identified on an iterative basis as revealed by the data during fieldwork (Fetterman 1989; Ozanne
and Hudson 1989). Each point is discussed, and support is provided by verbatim quotes from the
interviews; emergent themes are identified to clarify the meanings inherent in these experiences
for the homeless (Denzin 1989).

Acquiring Possessions

For the homeless, acquiring possessions involves activities that are markedly different from the
typical consumer in our society. First, many of the necessities of life (food, clothing, materials for
shelter, etc.) Are scavenged from the refuse of others rather than purchased. Second, access to
traditional outlets for products, such as supermarkets or restaurants, often is restricted for the
homeless because of financial and/or hygiene factors, dress requirements, and interpersonal
problems. Third, the homeless come in contact with nontraditional outlets for goods and services
(e.g., drug dealers), and many engage regularly in nonconventional acquisition (i.e., barter or
sharing with other homeless persons).

The Art of Scavenging. One of the primary means of acquiring possessions for the homeless is
scavenging. This activity involves the search for partially consumed products, with homeless persons as
secondary consumers. Sometimes items are used in their entirety by the primary consumer, and
nothing is left for a secondary consumer (e.g., a pair of shoes so worn that they provide little
protection from the cold or objects on the streets). However, many industrial and commercial
establishments as well as residential homes discard partially consumed products ranging from food
to clothing to such items as wood and cardboard that can be used in the construction of a shelter.
While the worth of these items varies with an individual's level of deprivation (e.g., at subzero
temperatures, even the smallest piece of clothing becomes valuable), the homeless persons we met
believe that others' "garbage" often has value.

You go look at [trash] cans today. People throw waste out-waste that's good waste-waste that can help
other people. [wm, thirties].

Most scavenging takes place in public garbage cans or dumpsters, which hold up to 20 cubic yards
of material. Uninformed observers may think that homeless persons aimlessly search such
receptacles, but those who scavenge are adept at acquiring useful items and employ developed
methods of selection and search to improve the likelihood of success in finding such goods as food
and clothing. For example, one key informant provided us with the following rule-of-thumb:

There're a lot of fast-food places and they all work pretty much the same. You find out when the place
closes, and you just go over there and climb in [the dumpster]. It's usually the same procedure. One bag
has all of the stuff from the kitchen, the other bags are all full of refuse paper. You get the heaviest bag
out, and that's the one with all the burgers in it, and fill up your sack. [wm, forties]

To avoid detection, scavengers often resort to late- night rummaging and random use of several
geo- graphically diverse sites. Others selectively scavenge dumpsters in areas where they have an
implicit agreement with various owners, managers, or workers. These people will carefully place
garbage bags containing whatever they consider to be of value to the homeless in the front of
the dumpsters at the same time every day or on the same day every week to facilitate retrieval
by the homeless.

Income Sources. While scavenging may provide the homeless with much of what they need for
survival, they seldom exist entirely outside of the cash-flow economy.5 A common misconception
by the public is that the homeless rely on the generosity of others through begging, charity, or
welfare to provide them with money. However, a majority of the homeless persons we encountered
were actively engaged in some form of traditional or nontraditional employment and were proud
of their independence from the welfare system. Some worked typical 40-hour-a-week jobs, while
others were employed sporadically as day laborers. However, nontraditional sources of income were
far more likely choices of the persons we met. One man summed up the situation this way.

People look at us like we are lounging around, but it's not easy-it's a job surviving out here!
We’re always working. At 4:30 in the mornings, I'm already out on the street. They don’t see us
because when they're sleeping, we’re working. [Wm, fifties]

One common method of making money is recycling. This source of income is consistent with the
outdoor, continuously mobile life-style of the homeless. Further, it requires little in the way of
equipment, personal presentation, or training. All one needs is a container (a plastic garbage bag or
grocery cart), a territory with reasonable potential, and a redemption center.

However, informants report that this business is in- tensely competitive, particularly when it comes
to returnable bottles and aluminum cans. Competition comes from retirees, lower-level employees
of businesses in the vicinity of their search, garbage handlers, and other homeless individuals.
Further, most retail outlets will accept only clean bottles or cans for brands that they sell. Since
homeless persons have limited access to water (particularly in the winter), these requirements
reduce the number of acceptable containers. A person would have to collect 280,000 bottles a
year to reach the U.S. Poverty level for a family of four. However, the most commonly reported
estimated amount for a day’s collecting is $6.

Because of the limitations on cans and bottles for income, homeless persons often turn to the
collection of scrap metal to earn money. The recycling process is about the same, but the value
of individual items often is significantly higher. There are three primary sources of metals accessible
to the homeless. One is cars that have been abandoned by either their owners or car thieves. These
usually are stripped of everything that is detachable (see photograph 2 in Fig. 1 for an example). One
can remove external equipment like hubcaps, trim, antennae, mirrors, and lights without entering
the vehicle. The car is then elevated (often with the abandoned car's jack), and the wheels and
tires are removed. Doors can be opened from the out- side with a flat piece of metal inserted
between the window glass and the metal frame. If this method is unsuccessful, a small window,
like the wing vent, can be broken with a brick or tire iron. Inside, the radio or tape player, seats,
and such accessories as mirrors and the lighter can be removed. Finally, the carpet, doors, and
hood are detached, and several engine parts, such as the radiator, battery, and air conditioner, are
broken loose. These pieces are then sold to parts dealers, scrap-metal yards, or directly to consumers
on the streets.

FIGURE 1-6
Although the entire car often will be gutted over time, the process usually involves several
individuals who specialize in the removal of certain parts. As the following quote suggests, they
each become adept at the acquisition and disposal of their particular specialty.

Anybody who is doing what I’ve being knows what to touch in a car and what not to touch. . . . [I] walk
around with my hammer or something, make sure there is no license plate or nothing on it to make
sure I won't get into no trouble. I take the radiator, air conditioner, the car battery, [or] anything that's
dealing with aluminum, copper, or brass.

A second source of scrap metal is abandoned buildings. When a building is deserted, the services are
physically cut. Boilers are decommissioned to dis- courage their being reconnected, electrical
supplies and water lines are severed, maintenance is discontinued, and doorways and windows are
sealed with concrete blocks. However, these precautions do little to dis- courage recyclers, who
use tire irons, crowbars, and other tools to gain entrance. Inside, virtually every- thing that can be
detached and carried is removed and sold. Recyclers rip open the walls to remove the pipes and
plumbing fixtures-usually only the bathtubs are left because they are too cumbersome to transport
to the scrap yards. The electrical wiring is extracted from the walls. The insulation is then burned off so
that it can be sold more easily. Elevator equipment and stairs are both valuable. Staircases are either
removed as a unit, or the marble steps and risers are taken, which leaves only the metal frame.
Finally, windows, doors, trim, door casings, and decorative woodwork are confisted. These items are
either sold or burned to pro- vide life-sustaining heat during the winter months.

The extraction of metal from buildings by homeless recyclers, similar to the automobile example, is
quite thorough. For example, one man provided us with the following process for removing the
coating from wire to facilitate recycling.

I only take wire if I see a whole stack-50 pounds or a hundred pounds. [To remove the plastic coating to
make it more valuable] I'll put it in a steel barrel, get a little gasoline, pour it in there, throw some wood
at the bot- tom of the can, start it, sprinkle the gas on it-once the Rubber come off I have clean copper. I
wait till it cool off, pour some water over it, turn the barrel upside down, take it all out, and then
compress the wire, and see how much I have. [bm,thirties]

A final source of scrap metal is dumpsters. Homeless recyclers look primarily for aluminum, copper,
lead, and brass. Excess metals of these types are discarded by window installers, building wreckers,
and remodelers. Success at scavenging such scraps requires an awareness of where such work is
taking place. For ex- ample, one of our informants specializes in the scrap discarded by
replacement-window installers. He searches during the day for sites where this kind of work is
occurring and raids the dumpster after the workmen leave, scrutinizing the debris for the largest
and most valuable pieces.

After the collection process ends, the difficult task of transporting these metals to recycling centers
begins. With bottles and cans, the number of redemption centers is larger because of the number of
retail establishments that must redeem them. However, recycling centers for scrap often are located
in remote areas of cities and suburbs. Since most homeless persons have no access to a functioning
automobile, they must find some way to carry the metals on foot. Thus, they often resort to
shopping carts and load them to or beyond their capacity (see photograph 1). This situation pro-
vides the uninformed observer with a picture of the homeless as eccentric hoarders of worthless
items. Nevertheless, as the following statement suggests, homeless individuals are patient in this
task.

Once it [the shopping cart] is filled up, then I'll go to the scrap yard. I'll walk there because it's not too
far. It's only 8-9 miles away from here. It don’t bother me – I’m used to it. Take my time - it’s no rush.
Every other block or so I stop to catch a breather. On the way, I might find a few more, you know, pound
of aluminum or whatever. [bm, forties]

Although recycling is probably the first choice of homeless persons seeking money, alternative
occupations exist. One is to be a "wiper” or, in street terminology, to “shine cars." Wipers make a
living by waiting at busy intersections for the traffic to stop, approaching cars, and proceeding to
clean the front windows. For this service, they receive a tip that may range from nothing to $100,
averaging about 25?-50. Some wipers report having good days that exceed several hundred dollars,
but an average day's tips are usually in the $50-$70 range. Many of these intersections are manned
by several individuals in an attempt to exclude others from their territory.

Nobody from nowhere else can come down here and shine no car. . . . They not allowed. If they do
some- thing, they [the police] blame it on us . . . so somebody is always here [to protect our turfl. [bm,
twenties]

The procedure is quite simple. The wiper uses a squeegee or a rag dipped in water (often from the
street) and attempts to clean the driver's side of the front window. Occasionally, they will spray
"cleaner" on the windshield so that the job will need to be finished. However, this occupation is
not without its hazards. Some wipers report being "stiffed" by "customers," threatened with
physical harm by angry drivers, and harassed by police, who write them tickets or confiscate their
equipment. Thus, to be successful, they must take certain precautions. One young wiper avoids trouble
by reading facial expressions to determine whether to approach a car.

You’ve got to really look at 'em. . . . If they smile, you shine their window. . . . You’ve gotta be a good
sales- man! [bm, thirties]

With regard to illegal activities, like drug dealing or prostitution, few of the homeless are involved in
such operations. Since these practices are often financially rewarding, most who engage in them
are able to find affordable housing. However, we did find homeless youths (under age 18) who
engaged in such activities, especially prostitution. One woman described a young man who lived in her
shantytown in the following way.
The kids that are homeless that are, let’s say, the age Of 13 and up-basically they survive by selling their
Bodies. We had a kid staying here, he was about 16, and all he knew was homosexuals-his world
revolved around homosexuality. [wf, forties]

Scavenging versus Buying. There are two interrelated reasons why homeless persons choose to
scavenge rather than purchase goods. The first is simple economics. Homeless persons are severely
constrained in their ability to afford food, clothing, and shelter, as well as other products. As the
previous subsection re- ported, the average daily wage of homeless persons who engage in some form
of nontraditional work is between $6 and $60-a small sum to provide the necessities of life.7 The
second involves their different perspective regarding what represents acceptable quality in the
items they consume. As stated earlier, the homeless often perceive value where others see garbage.
Several of our informants feel that they can get everything they need to survive by scavenging.

When money is available, homeless persons will splurge and treat themselves to something
special. At the top of the list often is a hot meal (especially in the winter), a favorite food that they
rarely have the opportunity to eat, or staples such as coffee and cigarettes. Unfortunately, informants
report that they or others in their situation may spend much of their limited funds on alcohol or
illegal drugs. One recovering addict provided the following perspective on the powerful impact drugs
have on the lives of his peers and how they quickly deplete all available financial resources.

Crack is so powerful that one shot will get you "ripped." If you got $500, you'll spend $500 on crack. If
you got $1,000, you'll spend $1,000. Whatever it take to get the Money to get it, that's what you're
going to do. If you're on welfare, they'll [drug dealers] give you credit. They’ll hold your ID card to cash
your checks so when check day comes, they return your ID and go with you to the check cashing place,
and once the check is cashed, you hand over the money to them. [bm, thirties]

Not surprisingly, the homeless often use nonconventional acquisition methods, such as barter or
sharing, to acquire products. It is not uncommon for one person to have a fire in a barrel, for
another to come over to cook food just purchased from a supermarket for the group, and for a
third to bring a bottle of in- expensive wine to share. These kinds of informal meetings take
place on a regular basis, with the same people typically in attendance.

Types of Possessions

The homeless need the same basic items as an average middle-class consumer-shelter, food,
clothing, and personal hygiene and health care products-and use a variety of tools to acquire and
transport these items. However, the ways the homeless fulfill their needs are often creative and
strikingly different from the remainder of society. The following discussion is organized by the major
categories of items consumed.

Shelter. Going from "housed" to "homeless" is rarely a sudden or unexpected event. Instead, it is
a process whereby an individual moves from a self-sufficient dwelling, such as an apartment, to
living with friends, relatives, or in government-controlled, temporary housing, to the streets. Each
homeless person we met had a different version of this same story. One woman in her thirties told us
that she had been living in an apartment but could no longer afford to pay the rent. She then
moved to an "SRO"-a welfare hotel that allowed her to rent a room on a continuous basis at a
relatively low rate. Unfortunately, because of an urban revitalization project, the hotel was
renovated and all of the residents were removed. It was at this point that she became homeless.
Other histories showed a similar spiral down the housing chain. One middle-aged male told us how he
was kicked out of his apartment by his girlfriend. He then lived in a public garage owned by his
brother but was forced out onto the streets after the patrons complained.

Governmental authorities as well as the general public often wonder why homeless individuals opt
for the streets over the shelters that are available in many communities. During the winter months,
the news- papers carry stories about homeless persons, characterized as "insane," who refuse to
leave the outdoors for the shelters. However, several visits to these establishments provided us with a
picture of an inhospitable alternative. Many of these facilities are overcrowded during their peak
usage months (i.e., the winter months) and consequently are poorly maintained. Privacy is
nonexistent, the smell of drugs and urine regularly fill the air, the staff at some facilities mistreat the
homeless, and some residents behave in a violent or threatening manner.

At some point, most homeless persons have tried shelters as an alternative to the streets, but few
intend to return. Their characterizations of the environment in these shelters are quite graphic.

[The shelters] they dirty, they filthy, and the people in employees] don't care, they treat you like there-
they thieves. I don't know who in there might have AIDS. They rob you, they devious, a lot of them have
lice. They smell, don't want to take a bath....They [shelter you nobody. They feel like this-if you in here,
you nobody cause you don't want to work. I'm not much better than them [shelter people], but at least I
keep myself from smelling. [bm, thirties]

Therefore, the homeless choose independent living and feel that survival, even under adverse
conditions, is possible.

Once you've really lived outside in the middle of winter-all winter long-it's no longer that scary thing.
When you get right down to it, a guy can roll up in a blanket in a snow bank and be warmer than a guy in
a house. It's just the idea of it-scares the hell out of you. You do what has to be done. You look back later
and say "Did I really do that?" and you really did. People have done that stuff for a thousand years. [wm,
forties]

A common perception is that the homeless occupy public spaces, such as train and subway stations,
shop- ping malls, or street grates, but our informants feel that they are more likely to be hassled
by police, security guards, teenagers, and so on in these places. Therefore, homeless persons
who are able choose to live more discretely, where their presence will go unnoticed or
unchallenged. Although the number of alternatives is rather large, the most common forms of
shelter are abandoned or unoccupied buildings, apartments, or automobiles, makeshift quarters
made out of scavenged construction materials, and partially protected, out-of-the-way places, such
as bridge abutments and tunnels.

Many homeless persons live in buildings that have been abandoned for years (see photograph 6).
Patient observation of these dwellings, which superficially seem devoid of life, often reveals
activity. Since occupancy is technically illegal, squatters take certain precautions to avoid being
discovered. Further, most buildings are infested with rats, and, if a ground-level entrance exists, wild
dogs may occupy the lower levels. Thus, for protection and to avoid detection, garbage is used
frequently to barricade inhabited areas, stairs leading to occupied rooms are removed at random,
and holes in the floor are covered with cardboard to booby trap entrances. We once watched a
man wall himself into a cellar using scavenged bricks.

Homeless persons may also search for an apartment that is between occupants in a populated
building or a house between owners or renters. Unlike abandoned structures, these facilities receive
attention, and, therefore, even greater secrecy is required. One informant summarized his strategy for
utilizing such dwellings as follows:

One of the number one [rules] of living like that is, even to another bum, you never tell him where you
live....The time that you do this [live in an unoccupied apartment in a populated building] you're always
taking notes of things. . . . You're "stealthily" when you come in and out. . . . The object is not to be
visible because the next thing you know you're getting run out. [wm, fifties]

Abandoned cars are also popular shelters, particularly when they are left on empty lots. Cardboard
often is used to cover windows and as a sleeping pad. Blocking the windows keeps the car cooler in
the summer, warmer in the winter, and allows for greater privacy. If the rear seat of the vehicle is
removed, the floor can be leveled and padded with layers of cardboard to shield a person from
exposed metal. Of course, many hazards exist with this type of shelter. Cars are more visible and
expose the homeless to potential victimization by gangs, criminals, and the police. Further, cars
may be towed by city authorities or dismantled by recyclers.
Makeshift quarters are the most elaborate of all the shelters used by the homeless. They may exist on
vacant lots, in shantytowns, or hidden away in the woods. To find materials for these structures, the
homeless will scavenge widely and are very resourceful at finding appropriate building materials.

It [his shelter] was out of the trash. A lot of stores threw out their rugs, so I went around and got those. I
had rugs over logs and plastic over the top of that to water- proof it. Gradually, inside I insulated with
cardboard boxes, trash bags, and leaves, and it was pretty tight. I didn't do it in one day-you fix it up and
you fix it up. [wm, forties]

One informant showed us a rather remarkable structure that was 3 feet high, 4 feet wide, and 8
feet long. The exterior was made of corrugated, translucent fiberglass, plastic garbage bags, and
5/8-inch plywood. The floor was made from an old wooden door covered with carpet. On this
particular day, it was raining heavily, but the interior remained dry.

Some of these structures contain multiple rooms. For instance, one woman gave us a tour of a dwelling
where she and her boyfriend lived. The structure had three compartments that included a bedroom and
separate, private areas for her cosmetics and his personal belongings. The entire arrangement was
intended to produce a homey atmosphere. The homeless often feel a strong attachment for these
shelters that they insist others who visit share.

Persons who live in places like bridge abutments and tunnels appear to be the least able to
fend for themselves (see photographs 3 and 4). Unfortunately, this environment is frequently the
most treacherous. While they have some protection from rain or snow, inhabitants are exposed to
the wind and cold, and they are more vulnerable to attacks by wild dogs or other people since they
are unable to isolate themselves. One of our informants who lived under a bridge became the
victim of bum burning-a sadistic practice of gangs of teenagers who pour gasoline on a sleeping
homeless person and then set him or her on ‘fire. The man we knew was ‘killed by an automobile
as he attempted to run across a highway to reach a nearby river to extinguish the fire.

Regardless of the type of shelter, maintaining a source of life-sustaining heat during the winter
months (known as the "death season") is an important challenge for the homeless. The most
common method of warming a shelter is to burn newspaper or wood they find while scavenging.
These are burned in buckets, bathtubs, sinks, stoves, old refrigerators, or any other fireproof
container that can hold a sufficient amount of material. One of our informants used a water heater
that someone had discarded as junk to develop a heating system for his shelter. He rigged it so that it
sucked air up from the bottom and sent smoke out the top of his shack.
However, the smoke and light of these fires can at- tract the attention of the police or the fire
department, with disastrous results. These authorities may destroy a shelter on the premise that it
poses a fire hazard to the surrounding community or wooded area. Also, the fire department may
extinguish the fire and simultaneously soak the entire living area, including bedding and clothing.
One homeless person described how such official action nearly ended a man’s life:

School called the fire department on the man [his shack is close to the school]. -He be burning wood out
here trying to stay warm. They wets up his bed because the fire is around there and everything [and] he
slept on it the same night.... This old man [who lived in the shack] comes to my car and asks me to give
him a cigarette, and how critical he be shaking from the cold. I said "George, take this drink." By the time
the cup gets to his mouth, half of whatever I give him was going out of the cup-he was shaking that
much. [bm,forties]

An alternative source of heat involves the use of electricity, and occupants of an abandoned
building may rely on stolen electricity. When a building is deserted, the utilities are cut at the
main junction box and the wiring is removed. Under most circumstances, however, power continues to
be supplied to this receptacle. The homeless may tap into the power line ahead of the fuse box and
use whatever wiring is avail- able to improvise a wiring harness that can generate electricity. This
form of “hot wiring" is not limited to buildings. Public street lamps are tapped in a similar manner by
inhabitants of both abandoned cars and makeshift quarters or shacks' (see photograph 5).

Once tapped, electricity is attached to an appliance such as a hot plate, electric frying pan, or
any other device that generates heat.8 However, this method of heating is not without its
dangers. The wiring is connected without the benefit of in-line fuses that normally protect
against overload. Uninsulated wire is used regularly, and several strands may be twisted together
to reach from a basement through a hall and into an apartment, or from a lamppost to a car
used for shelter. Without fuses, the power does not shut off until the wire burns, which is a fire
hazard.

All of the efforts described here-building or finding a shelter and devising a system to provide heat
in the winter-can be wiped out quickly. For example, one group of homeless persons who lived in
a shantytown told us that a representative from the "Department of Real Property/Vacant Lot Unit"
visited their com- munity and told them that their homes were to be demolished. Another
informant reported that the police asked city officials to cut off his source of electricity; they
subsequently placed a new metal plate over the site where he was tapping electricity from a
lamppost. Since he was then without heat, he was forced to search for a new location to live. Such
tactics by the authorities may lead to a further spiral down the housing chain from substandard
but adequate refuge to shelter less. Thus, impermanence characterizes these living conditions, and
homeless persons report moving many times over the course of several years.
Food. Not surprisingly, our informants eat every- thing from “roadkill” to sushi. And although food is
one of the products likely to be purchased when money is available, homeless persons secure much of
the food they consume through scavenging. The choice of “garbage" over soup kitchens or shelters for
nourishment bewilders many people in our society. However, consistent with their attitude toward
shelters for housing, the homeless persons we met felt that "shelter" food usually was inedible.

I don't like the shelters.... I said, "I'm not going to eat this mess, man; I cannot eat it." I asked the guy for
some salt, and he said, "You got any money?" . . . I went to another shelter, guess what they feed me-
army food! [bm, sixties]

Success in rummaging through dumpsters for food, like the scavenging tactics described earlier,
relies on planning. For instance, one informant told us that members of her community regularly
call take-out restaurants near closing time and place large orders. When no one comes for the food,
it generally is tossed in the trash. Then, after all of the employees have gone; it is retrieved by a member
of the group. Further, sifting through discarded -foodstuffs requires some method of distinguishing
between what is ingestible and what is not. Rules of thumb are common, and some homeless are aware
of current trends in healthful diet.

Open food that people have [partially eaten]-the least amount, you [can] get disease. The majority of
the time if it's cold or not warm, don't eat it. The best kind of food to eat today is fiber food. Too much
meat will hurt you; fish is better for you. [wm, thirties]

Once food is acquired, the homeless face problems with storage and preparation. Thus, they eat when
food is available but often go hungry. Occasionally, home- less persons use pirated electricity to run
a small refrigerator or a cooking device, such as a hot plate. Still others come up with resourceful
solutions. For example, a homeless male told us how he resolved a storage problem during the summer
months one year.

One time, with roadkills, I used to actually break into this place of business. They had a refrigerator in
there to keep all their sodas, all their stuff. So I could get in the window and stuff like a woodchuck in
the freezer, and hope that one guy would come in and say, "Oh Fred put a woodchuck in the freezer,"
and Fred would say, "Bill left a woodchuck in the freezer," and it would still be there when I went to get
it. [wm, fifties]

Preparation also has its difficulties. Without electricity, the homeless resort to cooking on an open fire.
However, as described earlier, the light and smoke from such a fire may attract unwanted attention.
Also, the smell of cooking or cooked food may lure a number of different animals, including rats,
mice, and wild dogs.
An additional problem involves the acquisition and storage of water. The homeless get water for
drinking or cooking from a number of sources, including fire hydrants and faucets at gas stations,
fast-food restaurants, churches, shelters, and public office buildings. However, they are restricted
by accessibility (e.g., in the case of fire hydrants a pipe wrench is required, and many retail
establishments restrict admission to paying customers) and by problems of transport. A steady and
adequate supply is difficult to maintain. Thus, the homeless must go without water on a regular basis
and sometimes resort to rainwater.

Clothing. Clearly, the primary purpose of clothing for the homeless is protection from the
elements, a concern of particular importance during the winter months. They have learned that
layers of material pro- vide the best protection, and they may supplement clothing with sleeping
bags, plastic trash bags, old curtains, and blankets, depending on their shelter and heating
arrangements, and may actually stuff leaves or rags between these layers.

Another purpose of clothing for the homeless is protection from attack. Thick material and a number
of layers reduce the impact of blows from a physical as- sault. Further, it decreases the vulnerability
of women to rape. For such reasons, homeless persons often wear large amounts of clothing even
during the summer months, which fuels the perception that they are men- tally unbalanced and
unable to make simple decisions concerning appropriate dress.

As with other basic commodities, the primary source of clothing is scavenging. However, clothing
drives and private charity also provide the homeless with items to wear. Further, although
everything one would normally wear is needed by the homeless, certain items are regarded
more highly than others. For example, sweatshirts, particularly the hooded variety, provide solid
protection and are easily removed and carried. Also, socks that are not too worn are difficult to
acquire and are needed. Finally, jackets or strongly constructed pants that have a number of deep
pockets that can be used for storage are valued.

Personal Hygiene and Health Care. Personal hygiene concerns among the homeless differ from those of
more typical members of our society for two primary reasons. First, the constant search for the basic
necessities of life and the everyday struggle for survival reduce the importance of hygiene to the
trivial.

When you're sitting out in a shack in the middle of winter, you got a fire going all the time; you're always
rolling around all the time in the dirt and leaves; you don't really think about how much you perspire
under your arms. There is nobody else to please there but yourself-you lose the [self-] consciousness.
[Wm, forties]
Second, restricted access to water reduces their ability to clean themselves or their clothing on a regular
basis. None of the abandoned buildings, automobiles, makeshift shacks, or any other alternative
shelter used by the homeless has indoor plumbing. Because they must carry water between the
sources of supply and their homes, the quantity available tends to be minimal. Further, consider
other unique difficulties associated with homelessness and cleanliness. The homeless need to wear all
of the clothing they own, particularly during the winter months. Thus, even if water, detergent,
and a washing basin or machine were available to them, they would have nothing to wear during
the cleaning process. Given these obstacles, the homeless clean themselves on an infrequent
basis, rarely wash their clothing, and urinate and defecate in the outdoors. Surprisingly, contrary to
their opinions about shelters in general, our informants felt that the shelters were useful for these
purposes, especially for the opportunity to wash oneself.

Health care presents homeless persons with a different set of obstacles. They typically suffer from a
wide variety of health-related problems caused by lack of food and water, substance abuse, and the
weather (i.e., extreme heat or cold). Nonetheless, because the home- less lack health insurance and
savings, most choose to ignore these problems until they are unable to function at all. At that point, the
first choice for health care is the emergency rooms of public hospitals or free clinics. However, a
surprising alternative choice by some of our informants was jail. One man provided us with the
following description:

Every time I feel myself going down, I feel depressed, I got nobody to talk to, I get aggravated by life, [I
go to jail]. . . . I do it deliberately [get arrested] just to go in there to get some decent rest and get
"cleaned up" [off drugs]. I'll just carry something I know I shouldn't be carrying. Unless I'm really, really
sick, I go to jail for help. They give me a thorough checkup-an AIDS test, you know, they give you a TB
test. They give you all type of body tests. [BM, thirties]

Tools. The homeless use tools primarily for acquiring possessions, transporting possessions, and
obtaining access to possessions. There are many ways that the homeless can transport
possessions, but the preferred method is by shopping cart (sometimes referred to as the "ship of
the ghetto"; see photograph 1). This "vehicle" can be used to carry large amounts of scavenged
materials to recycling centers or back to a shelter for personal consumption. However, an
alternative function of shopping carts involves security. Homeless persons, even if they are fortunate
enough to have some form of regular shelter, must carry the bulk of their belongings with them at
all times to avoid theft. Thus, they often heap these carts with a wide variety of items, including
books, clothing, food, other valued tools, and scavenged metals. One homeless man summarized the
dilemma faced by the homeless with the following example:

I knew a woman, she was an American Indian, and she used to take a shopping cart everywhere she
went. So if she walked into a restaurant to get a cup of coffee, boom, boom, in comes the shopping cart!
Is she trying to be visible? Is she trying to make a point? She just didn't know any better, I mean, she just
didn't want to lose that shopping cart. If you have anything you want secure, you keep it on you-you
don't leave it in a place. [Wm, fifties]

A second category of tools facilitates acquisition of possessions, particularly for recycling. Preferred
items include tire irons, ice picks, sledgehammers, screw- drivers, flashlights and candles, and
magnets. Tire irons are one of the most valued tools. They afford entrance to abandoned buildings
and can be used as a means of protection from wild animals or human intruders. Ice picks,
sledgehammers, and screwdrivers are useful in the removal of large metal parts from automobiles
(e.g., breaking the connecting bolts attached to a radiator) or metal pipes from abandoned
buildings. Flashlights as well as wax candles facilitate search at night and allow the homeless to see
in abandoned buildings that are sealed shut by the housing authorities. An interesting tool we
came across was the magnet. One informant described its use as follows:

Sometimes [in an abandoned building] they have the Old pipes-the brass pipes, the copper pipes coming
up through the ceiling or the side of the walls, or down in the basement. I know what to look for. I take
my magnet and, if it sticks, it means its iron, if it don't stick, the majority of the time it is either brass or
copper, so I know what I got. [BM, forties]

Community and Consumption

While some of the homeless persons we encountered had little or no regular contact with other people,
many were part of some type of support group. These groups band together for two fundamental
reasons. The first is protection. As mentioned earlier, the homeless are vulnerable to physical
assault and robbery from outsiders as well as harassment from governmental authorities.9 Thus,
homeless persons are wary of strangers, typically are unwilling to provide information regarding
the physical location of others, and often act to protect the possessions of members of their
community or shantytown.

The second reason involves the benefits derived from sharing. Rarely will a homeless person have all of
the necessities of life. Therefore, community consumption of available products improves the quality of
their lives. One woman described this kind of sharing in the following way:

One of the guys brings up ice when they come home; somebody else brings water up. If I come up here
and I’m hungry and somebody's cooking and [I say] "I’m hungry, can I have some of that?" and they say,
"No problem." [Bf, thirties]

The groups we encountered ranged from loosely aligned individuals who were aware of each
other's presence but interacted infrequently to thriving com- munities with daily communication.
The loosely connected groups can best be described as "shadow” com- munities. Conversations are
limited, and the basis of the relationship is shared assets. For instance, one per- son may possess a
pipe wrench that allows access to water from fire hydrants. In exchange for the water, another
person may provide food, liquor, or a place to store belongings. Even though they may exchange
resources only occasionally, homeless persons in these relationships are keenly aware of the
location and physical condition of each other and place a high value on these associations.

The other end of the spectrum includes enclaves of makeshift shelters referred to previously as
shanty- towns.'0 These communities are more explicitly social, and members regularly converse, sing
and play music, or even "work in the yard" together. Sharing in these enclaves is frequent, and
informants report a sense of duty, understanding, and caring toward the other members of their
group. Further, because of their greater visibility, the role of protecting one another and their
communal possessions increases in importance. Our informants who live in shantytowns re-
ported assaults by teenagers armed with bricks or by local residents who toss garbage at their
homes. Also, these "neighborhoods" may have rules that members must follow to remain within
their community.

One day my wife told him [a teenager who lived in their shantytown] to do something, and he said he
was no-body's maid, so we told him to leave. We felt sorry for him, but he was too hardheaded. All he
knew was "I got the money, and I don't give a shit," which is not right. [Wm, fifties]

INTERPRETATION AND

EMERGENT THEMES

To aid in the interpretation of this "thick description," the following emergent themes were developed.

The Homeless as a Nomadic Society

The homeless population studied in this investigation often survive by using the foraging mode of
subsistence typical of early peoples (see Lee 1979). This system is characterized by a reliance on
nature to pro- vide the necessities of life, mobility within an area large enough to provide sufficient
quantities of these items, and flexibility as changing opportunities are revealed within the
environment. Further, this approach may include living in "uninhabitable" locales and eating foods
that "outsiders" find repugnant. Such a survival strategy affects the character of possessions as well as
personal relationships with others.

With regard to possessions, the nomadic way of life of foraging societies strictly limits the
accumulation of wealth (Lee 1979). For people whose survival re- quires mobility, portability is a
major feature of the items that are retained. Thus, belongings generally are few, lightweight, made
from locally available materials, and multipurpose. The homeless clearly are aware of the limitations of
ownership imposed by mobility. Our informants expressed the need to carry valued items with
them at all times since they lacked safe storage facilities. In view of this requirement, the
homeless regularly wear bulky clothing with several pockets for holding items and acquire shopping
carts or laundry bins to transport larger possessions.
Further, as in foraging societies, the homeless scavenge most of their possessions from available
sources in their environment, acquiring goods that former owners have deemed worthless. From
abandoned buildings and cars to commercial and residential dumpsters, the homeless have learned
to find materials for shelter, clothing, recycling, and food. Such an approach to survival requires
flexibility because the same sources cannot be relied on to provide sustenance for any extended
period of time because of natural depletion and intervention from outsiders, including the police.
Thus, a continuous search for new opportunities and sources of supply is required to maintain a
relatively consistent level of resources.

Finally, some of the most valued possessions are tools that have several purposes and support
the for- aging mode of production. For example, the shopping cart facilitates storage, transportation,
and safety of belongings. Further, the tire iron provides access to abandoned buildings, aids in the
removal of valuable items from both buildings and automobiles, and can be used in self-defense.

In correspondence with early societies, exchange among homeless persons is best characterized by
reciprocity and has instrumental value that often acts to initiate or sustain social relationships.
Sahlins (1972) has provided a spectrum of reciprocities with extremes and a midpoint. Generalized
reciprocity refers to exchanges that are altruistic, such as voluntary food sharing among
kinspeople. Balanced reciprocity refers to the exchange of items that are perceived to be of
equivalent value without a time delay. Negative reciprocity is the attempt to get something for
nothing and includes such activities as gambling and theft.

The types of reciprocity practiced in foraging societies are a function of kinship distance and the
nature of the items to be exchanged (Lee 1979; Sahlins 1972). Reciprocity tends toward the
generalized variety with close friends and family and toward the negative variety with strangers.
However, even among close kinspeople, transactions may include a counter obligation, but the
expectation of reciprocation is indefinite and depends on what the initial giver needs as well as what
the initial receiver can afford when this need arises. Further, tools and other items with
instrumental value are more likely to be exchanged with balanced reciprocity, whereas food is
shared according to generalized reciprocity. The difference can be understood in terms of the
immediacy’ of the need and the dire con- sequences of going without. The following explanation by
Evans-Pritchard (quoted in Sahlins 1972, p. 210) provides the underlying rationale: "This habit of share
alike is easily understandable in a community where everyone is likely to find himself in
difficulties from time to time, for it is scarcity and not sufficiency that makes people generous,
since everybody is thereby assured against hunger. He who is in need today receives help from him
who may be in like need tomorrow."

Similar interactions took place among the homeless we observed and interviewed, especially in the
shantytowns, where a sense of community is more prevalent. In these "neighborhoods," a feeling
of kinship develops and results in protective, supportive, and sharing behaviors among individuals
within the enclave. Further, an understanding with neighbors and nearby residents evolves to
extend the network and protect against possible threats. Finally, suspicion thrives regarding
strangers, who often are perceived to be thieves or arsonists. As one young woman stated,

It's very fucking opportunistic! You suddenly start associating with people that are in tune with your
instincts. That are as aware of surviving at your particular stage. . It's like a pack or clan! There's codes
of behavior, regressing all the way to the most primitive behavior that must be adhered to or the
individual will suffer expulsion, which is a fear because he's already been expelled from society.
This is his nucleus.... This is protection [from outsiders]. [Wf, thirties]

Thus, where trust exists among the homeless, generalized reciprocity is the rule rather than the
exception. This is especially true for food. The homeless have experienced hunger on a regular or
occasional basis and can easily identify with those in need. Further, since many food items rarely
can be stored even in the most elaborate of homeless shelters, giving the excess above one's
individual needs to someone else has little cost. However, the shantytowns we observed extended
this mode of exchange beyond such necessities to include tools, storage, shelter, and personal
services that aid survival. As one informant told us,

You have to understand, these [the homeless persons in our group] are people from all walks of
life. One'll be a plumber, one'll be an electrician, one'll be an artist, one'll be a belly dancer.
Whatever it is, they will use their trades to help themselves and those around them. [Wf, twenties]

Generosity was more pronounced in times of extreme threat, such as particularly cold weather or
the imminent destruction of shelters by the authorities, which may have triggered a survival
instinct within the community and tended to increase the bonds among members (see both Lee
1979 and Sahlins 1972).

Self-Concept Development by the Homeless: Fighting a Deviant Label

Since the publication of The Presentation of Self in Everyday Life by Erving Goffman (1959),
researchers investigating the psychological and philosophical is- sues related to the self
increasingly have focused on social definitions and interpretations. For example, work in social
psychology suggests that our self-conceptions and resulting self-esteem are molded by
communications we receive from others, by comparisons with others, and by self-labeling that is
both socially learned and arbitrary (Mischel 1977). Thus, our sense of self is embedded in the
interactions and roles played within a society.
More specifically, the self-concept is an organized structure of various identities and attributes
and their evaluations, which are derived from an individual's reflexive, social, and symbolic
activities (Gecas 1982; Scheier and Carver 1980). Thus, a distinction can be made between the
content of self-conceptions (i.e., identities) and self-evaluations (i.e., self-esteem). Ac- cording to
Gecas (1982, p. 4), "Identity focuses on the meanings comprising the self as an object, gives
structure and content to the self-concept, and anchors the self to social systems. Self-esteem
deals with the evaluative and emotional dimensions of the self-concept."

With regard to self-concepts, most individuals are much less interested in reality testing than in
self-affirmation and self-protection (Becker 1971; Gecas 1982). For example, research suggests that
people often use the process of selectivity to distort their images of themselves (Rosenberg 1973).
This process affects both the sources of social influence and the social comparisons that are used
in the development of self-conception. Thus, individuals may seek out other persons who confirm
their self-identities and compare themselves with other groups to their own advantage.

Evidence consistent with this perspective suggests that some individuals resist deviant societal
labels and implied degradation and instead choose to fight back through repudiation or
modification of such labels (Rogers and Buffalo 1974). This form of adaptation may trigger the
self-efficacy motive, which causes individuals to seek control over the forces that influence their self-
conceptions (see Gecas 1982). Accordingly, individuals will seek personal power over the events
that affect and, therefore, define them by actively engaging their social and physical environments
(see Bandura 1977).

In our society, the label "homeless" is viewed as a public stigma and may cue collective
avoidance, ostracism, and isolation (see Goffman 1963). For ex- ample, Harper (1979, p. 25)
characterizes the literature in the social sciences on the "skid row man": "His lifestyle is not
integrated into the mainstream. His use of alcohol seems to be abnormal or diseased. He sleeps in the
open, or as a ward to the state in a mission; and his filthy clothes, messed hair, and offensive odor
mark him as a likely object of public disdain, scorn, or pity. As a 'deviant' his lifestyle has been
considered in terms of 'role-failure'-failure to integrate successfully into socially sanctioned places
in the social order." Snow and Anderson (1987, pp. 1339-1340) concur with this perspective and
feel that the homeless exist outside the role-based sources of self-esteem and human dignity that
most individuals in our society take for granted: "Their tattered and soiled clothes function as an
ever- present and readily perceivable 'role sign' or 'stigma symbol' that immediately draws attention
to them and sets them apart from others."

Given these impediments, the development of personal identities that lead to positive self-
conceptions is a difficult task for homeless persons. However, Snow and Anderson (1987) suggest
that the homeless often use identity talk to bolster their sense of self. For example, they found
that homeless persons use associational and institutional distancing to draw distinctions between
themselves and social identities inconsistent with their desired self-conceptions. These forms of
distancing were more pronounced among individuals who were not regular social service or shelter
users and who thus viewed themselves as more independent and resourceful.

Our findings support this conclusion.12 The home- less we investigated were individuals who lived
primarily outside the welfare system. These persons consistently denigrated the social services
available to the homeless and claim that reliance on such organizations reduces self-esteem.

It's [welfare and social services] geared to cause shame... We're talking about a society where status
quos and status symbols "emblem ate." These are the things that give a man a penis. This economic
prowess, social prowess. If he can't feed his family, if he can't get a job, he’s not a man. He's a dog. [Wf,
thirties]

Thus, our informants used associational (e.g., “I'm not like shelter users!") And institutional (e.g.,
“Welfare? Forget it!") distancing to bolster an image of themselves as persons living by their own
resources and abilities rather than under the control of these institutions. One informant
summarized these feelings in the following way:

A certain type of person likes welfare, another type of person doesn't like welfare. A guy that can do
something on his own doesn't think he should be taking money from them [the welfare office] when
there are people in wheelchairs that need that. You actually do have a responsibility-I mean I don't really
believe that you can bleed the state and get away with it. I think everybody just abuses that stuff
[welfare]-plus you abuse yourself when you don't never mind your own talents, your own
resourcefulness, and lean on somebody else. The state wants to run your life really when they get you
into something like that. It really rubs you wrong when you're into being fiercely independent. [Wm,
forties]

Our findings go beyond the work of Snow and Anderson (1987) to suggest that the activities as well
as the talk of the homeless are used to support their self- conceptions. Our informants participate
regularly in alternative work, such as recycling, to maintain a minimal standard of living. These activities
are often viewed with pride that is evident in their detailed discussions of how such work is
performed. Further, the homeless believe that by engaging in such activities, they are contributing
to the good of society rather than burdening the "system." For example, one man told us,

You feel good about it [recycling metals] because you don’t feel that you're [just making a buck], I mean
you're doing something that's worthwhile for society. [Wm, forties]
Further, an analogous sense of accomplishment is apparent in the scavenging activities that
provide the majority of the products they consume. Sources, techniques, and modes of adaptation
supply them with a feeling of resourcefulness that results in the belief that they can survive the many
perils of homeless life where others might fail.

The name of the game is survival on the street. There's a proper way and a wrong way. . . . I try to wait
patiently; I'm trying to alleviate a lot of problems I got. Maybe I have to go through what I go through for
a reason-on my own part. You go through life and you learn through the hard, rough [times], but [to
survive out here] you've got to learn to live off the land. [Wm, thirties]

Thus, the homeless persons we interviewed were able to maintain at least some self-esteem with
social and institutional distancing and with work to improve their standard of living and survival
potential under adverse conditions.

The Meaning of Possessions: A Comparison with Middle America

Recent research in consumer behavior contends that rising real incomes and an abundance of
consumer goods have resulted in a material outlook that construes possessions as an integral
part of self-identity (Belk 1985). While some scholars suggest that the satisfactions derived from
material pursuits are the result of self-deception (Wachtel 1983), few debate the importance of
belongings in our society. According to Belk (1989, p. 129), “As shown by studies asking American
adults and children the open-ended question ‘Who are you?’. . . possessions are prominently viewed
as part of self and are generally mentioned just after personal characteristics such as age and gender.
Possessions are part of the extended self in this society, while age and gender are more proximal."

This stream of research is influenced heavily by the work of Csikszentmihalyi and Rochberg-Halton (198
1), who focus on the symbolic value of possessions and their relationship to the self. While accepting
that the necessities of survival must be met first, these re- searchers believe that, for most members
of our society, "things" embody goals, make abilities apparent, and shape the identities of the
owners. Thus, people invest psychic energy in an object and channel part of them- selves into a
relationship with that object to the exclusion of other possibilities. To the extent that these
transactions with possessions result in the accomplishment of important goals or positive feedback,
they may strengthen the self and promote personal growth.

The consumer-behavior literature has focused attention on the importance of special possessions,
suggesting that such belongings are potentially growth promoting throughout a person's lifetime
(see Myers 1985). In a cross-cultural investigation, Wallendorf and Arnould (1988) found that
favorite objects enhance self-expression and promote differentiation and integration of an individual
within society. McCracken (1989) extends this perspective through his ethno- graphic investigation
of the North American home. He found that the creation of "homeyness" is one of the most
important goals in the transactions between people and their homes and leads to the inclusion
of such cherished objects as gifts, trophies, and family heirlooms. Surrounding oneself with
belongings in this fashion creates a buffer between oneself and the rest of the world. "The pragmatic
properties of homeyness give the individual a means by which to fashion their relationship with the
larger institutions of modern society. It lets them reckon with the intrusion of alien meanings
from the market place, the distracting competitive impulses of a mobile society, and the
unwelcome aesthetics of changing fashions. Homeyness helps the individual to mediate his or her
relationship with the larger world, refusing some of its influences, and transforming still others"
(McCracken 1989, p. 179).

Because belongings, including the home, hold such important meaning for individuals in our
society, the loss of material possessions is often viewed as a violation of the self (Belk 1988;
Wallendorf and Arnould 1988). Neal (1985) lovingly recounts the story of her grandmother, whose
dignity and control over her life were stripped from her by the gradual loss of her possessions as
she moved from her own house into a small room in a nursing home. Further, Lewis (1966) believes
that, in a society that values the accumulation of wealth and property, the lack of such possessions by
members of the lower socioeconomic strata may lead to feelings of fatalism, helplessness,
dependence, and inferiority. Such feelings arise because of the perceived improbability of their
achieving success in terms of the pre- vailing values and goals.

Clearly, these feelings exist among the homeless, who often have lost their possessions or had
them stolen and who cling to a precious few belongings that have symbolic value of times past.
Such items might include photographs, books, or mementos that remind them of happier days or
significant others. However, in accordance with Belk (1988), many homeless deal with such loss
through an attempt at self-restoration, particularly as it relates to shelter. The self-restoration process
usually involves three steps or stages. The first entails a perceptual change as the homeless person
modifies the habitual framework used to interpret what a "home" should be in a physical sense. The
standards for what constitutes an adequate shelter vary widely from culture to culture
(Csikszentmihalyi and Rochberg-Halton 1981). One homeless man showed us that such a
modification reduces many necessities to luxuries, which mirrors the needs of poorer societies:

[When you think of shelter] you don't think in terms of showers and things like this because you
[can] just jump in a stream or you get under where the canal passes over the stream and it leaks. You
know all these things- [like] having a bathroom-you don’t need those things. [Wm, fifties]
Second, while most members of our society develop a sense of pride from the purchase and
subsequent alteration of the home to meet their physical and psychic needs, the homeless may
develop similar feelings in the creation of their shelters. Although the living quarters are meager
by the standards of middle-class Americans, the homeless who build their own shelters are very
proud of their ingenuity.

I took two 2 X 4s and stuck 'em at the end of the bed as you can see. I tied ‘em on that end. I
took two more 2 X 4s, another 2 X 4 as a beam across, and tied it from one end to other. Then I
found me two doors and threw 'em on top, then I pulled the see-through plastic on top. And then I
threw a blanket and a cover and another piece of plastic, which is solid plastic, which is grained.
Okay, now it doesn’t drip! It can rain as hard as it wants. Right this moment, I’ll never get wet! [BM,
thirties]

Third, the meaning of the home, like that of a dream, does not lie in its manifest content but in its
underlying latent content (Csikszentmihalyi and Rochberg-Halton 1981). Thus, even for a shelter that
lacks many of the features common within middle-class society, the "home" still represents a
place where the owner or occupant sets the rules and is "king or queen of the castle."

This house is made out of wood and it's a shack, but it's our home and we respect our home, and
everybody who comes into the door is welcome to come in as long as they respect the house. [Wf,
forties]

Thus, through changes in perception, pride in creation of the shelter, and focus on the latent
content of what it means to have a "home," the homeless are able to cope with the loss of
possessions through a form of self-restoration.

CONCLUSION

Our research provides a sharp contrast with the work of Lewis (1966) by suggesting that homeless
persons take an active role in determining their life choices. These individuals confront their
challenging environments and engage regularly in endeavors designed to improve the quality of
their lives. This active role is particularly evident when the homeless are viewed as consumers. The
homeless employ unique adaptation strategies in their search for, and consumption of, goods and
services. These alternative consumer be- saviors allow them to survive, serve to restore meaning to
their lives, and bolster their sense of self.

As we stated, the homeless face a variety of restrictions that limit their ability to function as
ordinary consumers. Most prominent among these are hygiene, dress, and interpersonal problems,
as well as general economic distress. However, many overcome these obstacles through
nontraditional employment activities, scavenging, and the sharing of possessions through various forms
of community. Further, their active role in the acquisition as well as creation of many of their
possessions (e.g., the building of a shelter) tends to bolster their self-concept and increases their
self-esteem.

This perspective on the homeless does not imply that they are without need. Many go hungry frequently
and find themselves without shelter on a regular basis. Further, the most needy-the mentally ill and
the physically disabled-are the least likely to be able to devise and employ the survival strategies
described. Nonetheless, our research suggests that service providers to the homeless need to find
methods of support that allow individuals to maintain their independence and dignity through
nonconventional but self-reliant activities.

[Received January 1990. Revised June 1990.]