B. Write an APA Style Literature Review. ADD THE 2 NEW ARTICLES TO THE NEW RESEARCH DOCUMENT.

Progress in Human Geography 33(5) (2009) pp. 646–666

DOI: 10.1177/0309132508104995

Complexity not collapse: recasting the geographies of homelessness in a ‘punitive’ age

Geoffrey DeVerteuil,1* Jon May2 and Jürgen von Mahs3 1School of Geography, Highfi eld Campus, University of Southampton, Southampton SO17 1BJ, UK 2Department of Geography, Queen Mary, University of London, Mile End Road, London E1 4NS, UK 3Eugene Lang College, The New School for Liberal Arts, 66 West 12th Street, Room 904, New York, NY 10011, USA

Abstract: Over the past decade there has been a proliferation of work on homelessness by geo- graphers. Much of this has been framed by the desire to connect discussions of homelessness to wider debates around gentrifi cation, urban restructuring and the politics of public space. Though such work has been helpful in shifting discussions of homelessness into the mainstream geographical literature, too much of it remains narrowly framed within a US metric of knowledge and too closely focused upon the recent punitive turn in urban social policy. Here we advance instead a framework that recognizes the growing multiplicy of homeless geographies in recent years under policies that are better understood as multifaceted and ambivalent rather than only punitive.

Key words: homelessness, poverty management, punitive urban policies, revanchism, third sector.

*Author for correspondence. Email: [email protected]

I Introduction It is over a decade since Lois Takahashi (1996) reviewed homelessness research from a geographical perspective. She began by de- tailing the emergence then of a new home- less population (with growing numbers of women and children and people from minor- ity ethnic groups joining the traditional ranks of older, street homeless men), accounted for

its explosive growth, and reviewed the geo- graphically variable responses to the home- lessness crisis of the late 1980s and early 1990s. Rather less ground, she noted, had been covered on representations of home- lessness and homeless people. Such (mostly negative) representations were crucial, she argued, to a fuller understanding of the pro- cesses shaping a politics of rejection (see also

© The Author(s), 2009. Reprints and permissions: http://www.sagepub.co.uk/journalsPermissions.nav

Geoffrey DeVerteuil et al.: Complexity not collapse 647

Takahashi, 1997). Hence she developed a tripartite continuum of stigma, arguing that homeless people tend to be located within a harmful space of high culpability for their con- dition, non-productivity and dangerousness/ unpredictability. Crucially she also recog- nized that homeless people are not always presented thus, with some groups (women, or children, for example) as likely to engender sympathy as antipathy from policy-makers and members of the housed public. Thus Takahashi suggested that homeless people are best understood as occupying a varie- gated ‘space’ of stigma, and called upon geo- graphers to focus research on the ‘sites of representation’ that influence perceptions of homelessness and of homeless people so as to challenge that stigma.

Our goal in this paper is to extend Takahashi’s review of geographical work on homeless- ness into the twenty-fi rst century. We note its continued proliferation and identify key areas of research. But the review is framed by a desire to take forward Takahashi’s call for work that challenges conventional repre- sentations of homelessness. We do this by challenging the dominant means of repre- senting homelessness in the geographical literature itself.

Since the mid-1990s this literature – the majority of it emanating from the United States – has tended to be framed in punitive and legalistic terms that describe the progres- sive collapse of homeless spaces under the weight of measures that criminalize homeless survival tactics (eg, scavenging, panhandling) and clear homeless people from prime urban areas (eg, parks, railway stations, streets). Though we do not deny that the last decade has seen a number of punitive measures bearing down on homeless people, curtailing their access to public space and delimiting their ‘rights to the city’, we suggest that to frame homeless geographies exclusively in terms of ‘collapse’ is to ignore the increas- ingly varied and complex geographies of homelessness that characterize the contem- porary city.

We begin by rehearsing the better-known tenets of the punitive turn (section II), includ- ing its various overlapping strands, in roughly chronological order: considering responses to homelessness in the ‘carceral’, ‘revanchist’ and ‘post-justice’ city. In sections III and IV of the paper we introduce an alternative con- ceptualization of recent responses to the problems of homelessness – poverty man- agement. Rather than the simple collapse of homeless space, a poverty management perspective traces the recent proliferation of homeless spaces across the city and af- fords a more nuanced understanding of the processes shaping their emergence. In Part V we widen our remit to consider a broader geography of homelessness – tracing dif- ferent responses to the problems of home- lessness both within but also beyond the USA – and offer a framework through which these geographically variable responses to homelessness might be conceptualized. In the final part of the paper we return to Takahashi’s notion of a continuum or space of stigma to draw out another aspect of this variation: reviewing work which traces very different experiences of homelessness across a heterogeneous homeless population.

II The punitive framing of homelessness Here we identify three overlapping strands of the punitive framing of homelessness, and demonstrate how this framework has come to dominate understandings of homeless people and their geographies. Each strand – the carceral city, the revanchist city and the post-justice city – seeks to understand and critique socio-economic change in the late twentieth-century city in the United States and deploys homelessness (more specif- ically, homeless policies) as a crucial litmus test with which to understand the backlash in urban policy against the visibly poor since the 1980s (DeVerteuil, 2006). Taken to- gether, these strands paint an escalation of punitive measures against homeless people, beginning with the geographies of ‘malign neglect’, containment and concentration in

648 Progress in Human Geography 33(5)

the early to mid-1990s, and ending with more aggressive interventions that apparently relegate homeless people to an inexorably collapsing set of (public) spaces, the victims of anti-homeless ordinances, the destruction of Skid Rows, street sweeps and so forth.

1 The carceral city Much of the original inspiration for the pu- nitive framing of homelessness emanated from Chapter 4 of Mike Davis’ City of quartz (1990). In ‘Fortress LA’, Davis provided sting- ing evidence that fear of crime, disorder and the ‘other’, combined with a lack of com- mon civic culture and growing social and eco- nomic inequality, was transforming the built environment into a mosaic of fortifi ed en- claves, carceral spaces and ‘no-go’ zones. Drawing upon examples from Los Angeles, Davis detailed the simultaneous contain- ment of homeless people in a shrinking Skid Row (Davis, 1990: 232–36) and their in- creased exclusion from adjacent Downtown districts – through increased policing, and a hardening of the urban landscape itself (for example, ‘bum proof benches’ and sprinkler systems; p. 233). These policies, Davis sug- gested, had ‘turned the majority of homeless people into urban Bedouins … always fugitive and in motion, pressed between the offi cial policy of containment and the increasing sadism of downtown streets’ (p. 236).

While Davis’ chapter ranged widely, it was this degrading of homeless geographies, and the broader ‘malign neglect’ of home- less people (Wolch and Dear, 1993) that most captured the imagination of urban geo- graphers. Despite springing from examples in Los Angeles, the sense that homeless geo- graphies were becoming tightly contained elsewhere (and everywhere) resonated with other Leftist geographers and ‘dystopian urbanists’ (Merrifield, 2000), and led to a host of studies tracing apparently similar re- sponses to homelessness in cities across North America and western Europe (see, for ex- ample, Hoch, 1991; Sorkin, 1992; Zukin, 1995; Wolch, 1997; Metraux, 1999; Katz, 2001).

2 The revanchist city Two thousand fi ve hundred miles away, the emphasis on containment was being adapted to a distinctly New York context. Set against a dramatic early 1990s recession, faltering gen- trifi cation and a gathering backlash against street crime, panhandling and vagrancy, Neil Smith (1996a) captured an apparent shift in urban policy he characterized as ‘revanchist’. Revanchism is essentially the vengeful reclaiming of prime city spaces (eg, train stations, parks, sidewalks) by the (white) upper classes from street criminals, minorities, the poor and especially home- less people. Much of the empirical support for the revanchist city thesis is based on a few spectacularly punitive measures taken against homeless people in the late 1980s and early 1990s: the war against ‘squeegee men’, the rise of Business Improvement Districts (BIDs) and more intense policing of public space, and the destruction of home- less encampments in prime city spaces such as Tompkins Square, Penn Station and Times Square (see also Smith, 1992; Cresswell, 1996; Passaro, 1996; Dordick, 1997; Duneier, 1999).

Unlike Los Angeles – where the focus remained containment – in New York the response to homelessness was therefore more obviously interventionist, including the introduction of new ‘quality of life’ and ‘anti- camping’ ordinances designed to make sur- vival on the streets as diffi cult as possible for the city’s homeless (Smith, 1996a; 1996b; 1998). Furthermore, where in Los Angeles the battle against homeless people was mainly led by local businesses and the Community (sic) Redevelopment Agency, in New York Mayor Giuliani sought to secure the support of a wider public – representing the battle against ‘the homeless’ as a battle waged on behalf of the city’s (other) residents. Alongside in- creased policing there was therefore a vigor- ous media campaign reminding the housed public of the services available to homeless people in the city, urging them not to give to those on the streets, and seeking to legitimate

Geoffrey DeVerteuil et al.: Complexity not collapse 649

the forcible removal of homeless people to the city’s decrepit winter shelters (or, failing that, its asylum) as somehow in their own best in- terests (Smith, 1996a: 226; see also Mathieu, 1993). Given New York City’s liberal trad- itions, this shift in urban policy seemed both profound and profoundly unsettling – ‘the choice’, in the words of Smith, of ‘a narrow group of political and corporate leaders who really do constitute, with others, the ruling class’ (Smith, 1998: 16).

3 The post-justice city As with Davis’ work, the concept of urban revanchism quickly migrated from its ori- ginal context to emerge as a favored lens through which to chart a more punitive turn in homelessness and urban policies more gen- erally in numerous other western cities (on ‘revanchist’ urban policy, see Dangschat, 1997; Arapoglou, 2004; Coleman, 2004; Slater, 2004; Wyly and Hammel, 2005; Herbert and Brown, 2006; for the export of revanchist urban policy, see Smith, 2001; Swanson, 2007; for more critical accounts of the con- cept and its travels, see MacLeod, 2002; Brenner, 2004a; Hubbard, 2004; Uitermak and Duyvendak, 2008). Taken together, these accounts came to paint a particular picture of homelessness in the mid- to late 1990s. Homeless geographies were seen as ‘collapsing’, with intensifi ed policing and the proliferation of anti-homeless ordinances clearing homeless people from the prime spaces of the city even as gentrifi cation and redevelopment encroached ever further in to the marginal spaces of Skid Row.

For Don Mitchell (1997: 311), this (re)regula- tion and purifi cation of public space is leading to nothing less than the ‘annihilation of home- less people’ and their geographies: ‘creating a world in which a whole class of people sim- ply cannot be, entirely because they have no place to be’ (emphasis added). To Mitchell, recent developments suggest a move beyond revanchism towards what he has called the ‘post-justice city’ (2001: 81), a city ‘no longer defi ned by the struggle for social justice …

[but rather the] question of the best way to exterminate homeless people’.

Though building on Smith (1992; 1996b), Mitchell extends earlier work in this vein in at least three ways. First, while for Davis (1990) and Smith (1996a) attempts to clear home- less people from the streets seem mainly to have been undertaken at the behest of local businesses and developers, for Mitchell these attempts speak to a far broader logic. Rendering more explicit the scalar politics evi- dent in such campaigns, Mitchell positions attempts to clear the streets as lying at the heart of a new concern among city man- agers to bolster their city’s image in an era of intense interurban competition: securing the local conditions (of an attractive, ‘livable’ cityscape) demanded by international capital (Mitchell, 2003: 163–67; see also Mair, 1986). Second, extending Smith’s analysis of anti- homeless ordinances, Mitchell traces their growing ubiquity – noting their emergence in numerous cities across the USA, includ- ing even apparently liberal bastions such as San Francisco, and launching a damning critique of the effects of such laws on homeless people’s wider ‘rights to the city’ and to citizenship (Mitchell, 1997; 1998a; 1998b; 2001; 2003; 2005). Third, in contrast to both Davis and Smith, Mitchell pays rather more attention to the role of local welfare agencies in tackling the problems of homelessness, seeing initiatives like San Francisco’s Matrix Program (an aggressive ‘outreach’ program funded by the city and designed to move people off the streets and in to shelters provided by local third- sector organizations) as an ill-concealed (and largely unsuccessful) attempt to legitimate street clearance campaigns and further dis- cipline those homeless people who ‘refuse’ assistance (Mitchell, 2003: 180–81).

Mitchell’s work has been enormously infl uential, framing discussions not only of homelessness but of legal geographies and wider debates around rights to public space and to the city (see, for example, Fitzpatrick, 2001; Law, 2001; Lee and Farrell, 2003;

650 Progress in Human Geography 33(5)

Blomley, 2006; 2007a; 2007b; Cowan, 2006; Cameron, 2007; Staeheli and Mitchell, 2007). The post-justice city – and its emphasis upon public space law – has become the frame within which discussions of homelessness by geographers now tend to proceed (see, for example, Law, 2001; Brinegar 2003; Collins and Blomley, 2003; Lees, 2005; Laurenson and Collins, 2006; 2007; DelCasino and Jocoy, 2008). At the 2007 annual meeting of the Association of American Geographers, for example, no fewer than three sessions were convened around the specifi c theme of rights, space and homelessness (Blomley and Klodawsky, 2007a; 2007b; 2007c).

III Beyond the punitive frame? That recent years have seen an increasing array of punitive measures bearing down on homeless people and on their use of public space in particular is undeniable. Echoing developments in the United States, the past decade has seen a proliferation of new initia- tives in both the United Kingdom (Anti-Social Behaviour Orders, Public Drinking Exclu- sion Zones, and Town Centre Partnerships, for example) (Coleman, 2004; Johnsen and Fitzpatrick, 2007) and elsewhere (Laurenson and Collins, 2006) that are now being used to restrict the movements of street homeless people. Nonetheless, at least three ques- tions need to be asked about the punitive framing that now dominates discussions of homelessness in the discipline.

First, who are ‘the homeless’? Although those working within the punitive frame have much to say about discursive constructions of homelessness, about recent attempts to curtail the rights of homeless people and even – if less frequently – about attempts to resist such changes, they tell us very little about homeless people themselves and about how such changes might be experienced by different homeless people. Indeed, at times, those working within the punitive frame stray dangerously close to a portrayal of the kind of universal homeless subject found in media

accounts (on which these authors so often draw), accounts in which the experience of homelessness tends to be reduced to the experiences of only street homeless men (May, 2003; for a rather different reading of the construction of a universal homeless subject, see DelCasino and Jocoy, 2008).

Yet we know that the homeless popula- tion is in fact remarkably heterogeneous. Just as some people are far more likely to become homeless than others, so too experi- ences of homelessness – including whether or not a person is liable to be among the (rela- tively small) numbers of street homeless people (rather than the much larger shelter population) swept up in recent street clear- ance campaigns – differs widely according to age, gender and ethnicity (von Mahs, 2005; May, 2008). Even on the streets, experiences of homelessness differ signifi cantly between men and women (May et al., 2007) and be- tween different subgroups within the wider street homeless population (Johnsen et al., 2005a). Rather than continue to talk of ‘the’ homeless, we would suggest the need to develop a more nuanced understanding of a varied homeless population and therefore of the quite different experiences of home- lessness within the contemporary city.

Second, are the geographies of home- lessness really everywhere the same? In the tendency to apply a punitive frame to read- ings of homelessness, we can trace the same debilitating tendencies noted by Brenner (2004b) in his critique of paradigmatic think- ing in urban studies more generally. As the punitive frame becomes what Hetherington (2001) would refer to as an ‘obligatory point of passage’ in discussions of homelessness, the rush to identify yet more examples obscures all other responses to homelessness, even in cities where the punitive frame cannot hope to capture the complexities of these responses (see Laurenson and Collins, 2006). In a world of intense interurban competition and rapid policy transfer the fact that similar, and sim- ilarly punitive, responses to homelessness are

Geoffrey DeVerteuil et al.: Complexity not collapse 651

evident in a number of Western cities is not altogether surprising (on policy transfer in this and related fi elds, see Hubbard, 2004). Rather than document more examples of the same, it would seem more useful to try to understand why it is that problems of home- lessness take different forms in different places, and why responses to these problems likewise differ.

Third, who precisely has banished home- less people from the streets and – if banished – where have they gone? Notwithstanding the continued focus on the agents of contain- ment and control – rather than on the subjects of that control – those working within the punitive frame often provide only the vagu- est picture of precisely who is responsible for this recent shift in homelessness policy. Most often the fi nger is pointed at the ‘local state’ which, with the support of an apparently compliant media and police, seems set on clearing the city of homeless people on be- half of local businesses, developers or ‘inter- national capital’ (see, for example, Smith, 1992; 1996b; Mitchell, 1997) Such accounts offer an inadequate reading of both capital and especially of the state, and of the relation- ships between capital, the state and other actors. We suggest instead the need to open up our theorizations of ‘the state’ (Peck, 2001) to take better account of the tensions between different arms of the local state (between ‘regeneration’ agencies and social service departments, for example, each of whom may be pulling in rather different directions), between the local and regional (or federal) state, as well as between the state and other agencies with a role to play in responding to the problems of home- lessness. The latter include both business interests and the police but also (and espe- cially) the third-sector organizations respon- sible for the delivery of homeless services in most North American, western European and Australasian cities (von Mahs, 2005; DeVerteuil, 2006; Laurenson and Collins, 2006; May et al., 2006). In many instances it is possible to trace a far more accommodative, or at least ambivalent, response to homelessness

than has hitherto been recognized: partly because the state cannot in fact be reduced to the handmaiden of capital, and partly – as early work on urban growth coalitions made plain (Leitner, 1992) – because the coalitions through which homeless policies unfold are far more diverse (and far more fragile) than is implied in the punitive literature. Such diver- sity tends in turn to lead to a variety of policy responses, sometimes pulling in different directions, in the same place.

Once this variety of response is recognized (see also Law, 2001), far from collapsing, the geographies of homelessness proliferate. Rather than focus only on the restrictions imposed upon homeless people’s access to public space, then, the more pressing need is to examine the changing constitution of those other spaces in which the majority of home- less people spend most of their time: shelters, refuges, prisons, drop-in centers and rooming houses. It is into these spaces that homeless people are ‘disappearing’ (cf. Smith, 1998: 5): as the past decade or so has seen not a collapse but radical expansion and disarticulation of spaces of homelessness across the city.

Moving beyond critique, in the remaining sections we elaborate upon each of these issues in turn. In section IV we turn to an alter- native conceptualization of recent responses to homelessness – poverty management – and outline the more complex and varied geography of homelessness this concept brings to our attention. In section V we broaden this geography and introduce a new framework through which we might con- ceptualize variations in both the nature and extent of homelessness, and responses to it, in different places, and review some recent work by geographers on homelessness in different cities both within but also beyond the USA. In section VI we return to the ques- tion of heterogeneity, reviewing recent work by geographers that offers a more nuanced account of differences within the homeless population and that attempts to say some- thing about homeless people’s own experi- ences of the city – including their responses to a more punitive urban policy regime.

652 Progress in Human Geography 33(5)

IV Reconceptualizing responses to homelessness: containment, control – and care? Here we examine recent social policy re- sponses to homelessness that articulate more ambivalent, possibly even accommodative responses to the problems of homelessness that belie any single-minded punitive turn in urban policy. Rather than consider only how the state has sought to contain and control ‘the homeless’, we approach instead the ways in which the state, and others, may also be seeking to ‘care’ for homeless people – even if, as we note below, the line between ‘care’ and ‘control’ is not always easy to de- termine (Piven and Cloward, 1993).

One way to conceptualize these more varied responses to homelessness is to turn to the concept of poverty management (Wolch and DeVerteuil, 2001; DeVerteuil, 2003; 2006). Drawing on concepts of the public city, urban managerialism, governmentality, abeyance theory and landscapes of care, ‘poverty management’ refers to the creation of spatial and temporal structures designed to regulate and manage the spillover costs associated with so-called disruptive popula- tions. Within a poverty management frame, the state – alongside other institutions and elites – promulgates specific poverty man- agement techniques that range from the supportive (eg, affordable housing) to the ambivalent (eg, allowing the third sector to set the homeless agenda) to the punitive (eg, anti-homeless ordinances). These strategies translate into specifi c sites of management that vary over time – from nineteenth-century asylums to the contemporary homeless shel- ter. With the dramatic growth of homeless- ness in the heart of a number of (upgrading) American cities in the 1980s, a backlash against this growth (perceived by many in terms of a crisis of social control) was inevit- able; leading to the development of a new regime of poverty management (Wolch and DeVerteuil, 2001).

The goal of this new regime was to reduce homeless visibility through both ‘entrenched

community opposition [to shelters and other services] and … anti-vagrancy laws’ but also, and crucially, ‘re-institutionalization and cir- culation’ (DeVerteuil, 2006: 119). As a result, though homeless people in large American cities may now find their access to prime public space ever more restricted, they are also increasingly to be found regularly moving between an array of unrelated, inadvertent, informal and inappropriate settings – including shelters and SRO hotels, but also jails, prisons, hospitals, rehabilitation centers and recovery homes (DeVerteuil, 2003). This reinstitu- tionalization and circulation of homeless people thus introduces a far more complex geography of homelessness than has hitherto been traced. Far from any collapse, there has in fact been a remarkable proliferation of homeless spaces over recent years, includ- ing the considerable expansion of abeyance, ambiguous and interstitial structures that accommodate and support homeless people (Laws, 1992; Williams, 1996; Knowles, 2000; Burt et al., 2001; Datta, 2005; Lobao and Murray, 2005).

In a study of the Los Angeles shelter sys- tem in the late 1990s, for example, DeVerteuil (2006) found that, far from shrinking, the number of shelter beds available to homeless people was in fact growing – with the larg- est single concentration of beds remaining in the Downtown (Skid Row) core. In Britain, too, the number of emergency shelters – 95% of which are provided by third-sector organizations supported by a combination of local and central state funding and public donation – has grown dramatically in recent years, at a time when sympathy for homeless people is apparently faltering (May et al., 2006).

An increase in the provision of emergency shelters is not necessarily incompatible with arguments of containment and control. For the local state, shelters may provide an effective way of shifting homeless people off the street – thus rendering the problems of homelessness less visible – at minimal cost to the public purse or a politician’s image.

Geoffrey DeVerteuil et al.: Complexity not collapse 653

Even so, the funding of emergency beds arti- culates a more complex response to the prob- lems of homelessness by the local state than simple ‘extermination’.

In fact, the majority of such shelters (as well as other services for homeless people such as soup kitchens and drop-in centers) are provided by third-sector organizations. Although often reliant upon local and central state funding, these organizations also tend to remain heavily dependent upon public donation to fund and staff their services. That such services have increased in recent years suggests that ‘revenge’ is not the only thought running through the minds of the public when confronted by homeless people. Indeed, we might point instead to deep- rooted and still powerful feelings of charity towards homeless people: feelings we have conceptualized elsewhere as an ‘urge to care’ (May, 2008), and which are apparent even in the USA (Link et al., 1995). Such an urge is most evident, perhaps, in the efforts of the tens of thousands of volunteers who continue to give their time and energies in service to homeless people, but also in the continued failure on both sides of the Atlantic of ‘diverted giving schemes’: with members of the public still offering money to those on the streets (rather than to local charities) despite the protestations of politicians and policy-makers that to give directly works only to support a damaging ‘street lifestyle’ (Dean, 1999).

Geographers and others have been re- markably slow to look inside these service spaces so as to make an assessment as to how best they might be conceptualized. The picture that emerges from the few studies we have is diverse. Certainly many third-sector organizations continue to provide only the most basic facilities. Many shelters provide residents with few comforts and little secur- ity. In some conditions are so bad, and the con- ditions of entry so restrictive (ranging from early morning exits, evening curfews and a complete ban on alcohol), that some home- less people prefer to take their chances on

the streets rather than risk the violence that can be a feature of these spaces, or submit to the control of shelter staff (see, for example, May et al., 2007).

At the other end of the spectrum, for Williams (1996) the highly intrusive personal service programs now in place in many ‘high support’ shelters are evidence of a profes- sional social services approach within which it is ‘homeless people themselves, rather than poverty, unemployment, or low-income housing shortages [that] are the “problem” to be addressed and corrected’ (Williams, 1996: 75) (for similar arguments, see Veness, 1994; Fopp, 2002; DeVerteuil, 2004). In light of such programs, for Williams, the question to be asked is not whether current approaches to homelessness are characterized by attempts to take control of, or to care for, homeless peo- ple, but to what extent the provision of wel- fare is also and always an attempt to impose control over welfare recipients. Others have found both these much more basic services and services similar to those encountered by Williams (1996) alongside shelters offering good-quality accommodation, security of tenure and high levels of support (for those that wish to access it) but without the kind of intrusive individual case management pro- grams described by Williams, all in the same city (May et al., 2006).

The evidence thus suggests that, even while other forms of state welfare important to homeless people (most notably afford- able housing programs) continue to decline, the provision of emergency services such as shelters, drop-in centers or soup kitchens is expanding in many cities (Johnsen et al., 2005a; 2005b; May et al., 2006). As those providing these services would be the fi rst to admit, emergency shelters and drop-ins do not hold the solution to homelessness, but nor are they simply a ‘sticking plaster’ or a tool of containment. Many such organ- izations now provide a stepping stone for their clients to move off the streets more per- manently (in the form of transitional housing, or tenancy support programs, for example)

654 Progress in Human Geography 33(5)

and, even if their approaches differ, all are driven by a genuine desire to help homeless people (MacLeod, 2002; Cloke et al., 2005; 2007a). What is clear is that the nature, quality and supply of such services is highly uneven. In this regard they refl ect the uneven geographies of voluntarism more generally (Milligan and Fyfe, 2004; Wolch, 2008). Also, as with other voluntary services, their ability to affect a longer-term solution to homeless- ness remains constrained by shortages (of affordable housing, or drug and alcohol rehab programs, for example) elsewhere in the wel- fare system (May et al., 2006).

A poverty management perspective there- fore implies that the precise balance between a more obviously punitive and more obviously accommodative response to homelessness will vary signifi cantly over space (and time), depending upon the presence of these kinds of organizations, the motivations of the dif- ferent agencies responsible for formulating a response to homelessness, and the balance of power between these agencies in any one place. For example, May et al. (2005) report on a wide variety of outcomes following at- tempts by the British government to impose a more coherent framework for dealing with problems of street homelessness in its Home- lessness Action Programme (1999–2002). In some cities, the program was quickly em- braced by local authorities keen to attract the extra funding made available by central government. In these places, the local state quickly bought pressure to bear on the third- sector organizations they funded to abide by the key tenets of the program – more aggres- sive ‘outreach’ programs, a clampdown on begging and a stricter rationing of emergency beds – so that future central government funding might be secured. In those same cities, however, a number of other voluntary-sector organizations – notably those providing on- street services for homeless people, and those with more diverse funding streams – remained fi rmly ‘outside the loop’, frustrating the efforts of the local state to impose a more coordin- ated (and more draconian) response to the

problems of street homelessness. Elsewhere, local authorities shied away from making an application for Homelessness Action Pro- gramme funding altogether: concerned that to apply to funding was to ‘admit to’ a problem that could damage the city’s image, even if the cost of not applying was fewer resources to deal with that problem. As a result, while the third-sector organizations in those cities remained starved of resources, they also continued to enjoy much greater freedom to determine their own policies and procedures (including to whom to offer assistance, and in what form) than their colleagues elsewhere (Cloke et al., 2007b; see also Law, 2001, for similarly varied responses to homelessness in Los Angeles County).

Another obvious weakness of work pro- ceeding within the punitive frame is the ten- dency to attribute unambiguous (and usually wholly negative) motivations to those thought to have led the campaign against homeless people in recent years. Yet such motivations are not always so easy to determine. For ex- ample, each year major corporations and banks in the United States, Britain and else- where donate food and money to numerous emergency programs, begging the question of precisely where a ‘revanchist’ or ‘post-justice’ approach sits with the mantra of ‘social re- sponsibility’ and ‘corporate citizenship’ now publicly advocated by a number of corpor- ations. Such donations are crucial to the survival of many services. That these busi- nesses donate to ‘off-street’ rather than ‘on- street’ services may be evidence that donors would prefer these programs to operate be- hind closed doors (and out of sight), but it is hardly in keeping with the idea of an attempt by business to ‘exterminate’ homeless people (Johnsen et al., 2005a; 2005b; cf. Mitchell, 2001: 81).

Nor are the ‘anti-homeless’ coalitions iden- tifi ed in the revanchist/post-justice city litera- tures always as robust as is often assumed. For example, a recent attempt by the London Assembly (the body representing London’s 32 local authorities) to stop soup kitchens

Geoffrey DeVerteuil et al.: Complexity not collapse 655

from operating on London’s streets was aban- doned after vigorous protest from a number of third-sector organizations – many of which member authorities would have to continue working with to deliver other aspects of their social service programs (Evans, 2007). Pressure on these coalitions can also come from the most unlikely sources; Johnsen and Fitpatrick (2007) report on the inability of some local authorities in England to increase the number of Anti-Social Behaviour Orders served on those who persistently begged in their town centers following the refusal of local police to enforce these orders unless tied to a more obvious and well-publicized support package for the individuals concerned.

All of this suggests that recent responses to homelessness have been far more complex, and more varied, than those writing within the punitive frame would have us believe. Rather than continue to focus only upon the ‘purifi cation’ of public space and the banish- ment of homeless people from the city, we need instead to recognize the increasingly complex, sometimes incoherent, geographies of homelessness evident in the contempor- ary city. In the following section we extend this geography, providing a new framework through which we might build a better under- standing of why responses to homelessness differ in different places.

V Beyond a US metric of knowledge: a new framework for understanding the production of and responses to homelessness It is clear that whatever the impression given by those writing within the punitive frame the imposition of anti-homeless legislation, street sweeps and so forth is far more pre- valent in some cities (notably, those with a heavy reliance upon the fi nancial and creative industries, tourism and the convention trade) than in others (Wyly and Hammel, 2005; May, 2008). Indeed, one of the more frustrat- ing aspects of recent work by geographers on homelessness is the tendency to proceed as though the problems of homelessness – and

responses to those problems – are implicitly everywhere the same: or, more accurately, perhaps, the same as in New York City or Los Angeles. Yet we know from re- search that even these two cities are quite different in their approaches to home- lessness and have different homeless geo- graphies. New York City is one of the few American cities legally obliged to provide public shelters for homeless people (Gaber, 1996). Combined with other measures – an exten- sive transitional and permanent affordable housing program, for example – the result is considerably less on-street homelessness than in the (less populous) City of Los Angeles (Inter-University Consortium Against Home- lessness, 2007). In contrast, like most Sunbelt cities, Los Angeles provides only a minimum of state funding to third-sector organizations operating within its city boundaries, creating a ‘laissez-faire system of indirect support, contracting-out and service fragmentation, insuffi cient capacity, and low degrees of inter- connectedness’ across the local shelter system (DeVerteuil, 2006: 112). As a result, a critical lack of shelter beds renders the problems of homelessness much more visible.

This tendency to proceed as though devel- opments in other places – and other countries especially – are unfolding along the lines described in the USA (read Los Angeles or New York City) is deeply problematic. This problem may, in part, be one of volume. It is still easier to find work by geographers on homelessness in the United States than on homelessness elsewhere. Yet, even when research ventures beyond the United States, the tendency remains to focus upon aspects of homelessness in those places which can most easily be related to a US – or, more ac- curately, punitive – metric. Hence, research has focused upon the experiences of the street homeless population (rather than on other homeless groups), on homeless people’s experiences of public space (rather than their experiences of other spaces of homelessness), and on questions of control and containment rather than on other, more accommodative,

656 Progress in Human Geography 33(5)

responses to the problems of homelessness. The result is the unintended impression that the geographies of homelessness are much the same in cities otherwise as different as Athens (Arapoglou, 2004), Liverpool (Coleman, 2004) and Hamburg (Dangschat, 1997) as they are in New York City and Los Angeles.

In fact there are good reasons to suppose that the problems of homelessness and responses to those problems will take differ- ent forms in different places. With few com- parative studies on which to draw (though see Glasser, 1994; Marr, 1997; von Mahs, 2005; Minnery and Greenhalgh, 2007), here we go beyond a critique of the punitive frame to offer our own explanation of why this might be so, and a potential framework for future comparative research (for an earlier formu- lation of this framework, see May, 2004). While we can trace a growing convergence in responses to homelessness in a number of western European, North American and Australasian cities, to see in this convergence the emergence of a single – and universal – response to the problems of homelessness in the contemporary western city is problem- atic for at least four reasons. First, it is now widely accepted that both the size and char- acteristics of the homeless population change in response to broader changes in housing and labor markets and in state welfare regimes that impact especially hard on particular groups (Wolch, 1996). Although the direction of change across these fi elds has been con- sistent in recent years (tied to the broader neoliberalization of western economies and welfare) such restructuring remains locally contingent and path-dependent (Larner, 2000; Harvey, 2005; von Mahs, 2005). As such, different pathways are apt to produce different levels and forms of homelessness in different places.

Second, just as state welfare regimes exert an important influence on the production of poverty and homelessness (helping to determine both the size and nature of the homeless population) (see Wacquant, 1999), so they also help determine responses to

homelessness in that place. Others have identifi ed variations in the nature and extent of state welfare provision in different countries – according to whether these countries fol- low a broadly ‘social democratic’ (ie, Sweden, Norway), ‘conservative’ (ie, Germany, France) or ‘liberal’ (ie, USA, UK) welfare regime (Esping-Andersen, 1990; Daly, 1999; von Mahs, 2005). So too we might identify differ- ent responses to homelessness within these broader regimes, with some countries (ie, ‘social democratic’ Scandinavian countries) more generous than others (ie, ‘liberal’ Anglo- Saxon countries) in the help they extend to homeless people through mainstream wel- fare services, or in the groups they extend such services to.

Such arguments align with more recent studies questioning the degree of convergence evident in different western welfare state regimes in an era of neoliberal restructuring, and which point instead to the more limited inroads that a US-inspired approach to welfare prevision has made in a number of western European states (Alcock and Craig, 2001; Huber and Stephens, 2001). In the current case, however, the nature of mainstream wel- fare provision may in turn infl uence policy responses more directly concerned with the problems of homelessness – helping to de- termine whether such responses are broadly accommodative (the increased provision of emergency shelters), or more punitive (street sweeps and anti-homeless ordinances). These responses are liable to be further complicated by the role that third-sector organizations play in homeless service provision at the local level, adding further variation within the na- tional or regional context.

Third, we might point to the role that less easily quantifiable – but nonetheless important – differences in the cultural sig- nifi cation of homelessness (whether or not problems of homelessness are recognized and, if so, how constructed) play in shaping re- sponses to homelessness in different places. Such differences are as likely to play out at the local level (or across an urban–rural divide)

Geoffrey DeVerteuil et al.: Complexity not collapse 657

as they are at the regional or national level (Cloke et al., 2000; 2007b). Fourth, we might point to the role that a number of other locally specifi c factors play in both the pro- duction of and responses to homelessness: whether the cost and availability of illegal drugs (with the greater availability and falling cost of heroin especially a key factor in the rapid rise of street homelessness in Britain in recent years, for example; Neale, 2001), or the urban morphology (helping to deter- mine the relative visibility of street homeless people within a city; see, for example, Snow and Mulcahy, 2001).

Precisely because they operate relation- ally, the specifi c ways in which these factors shape responses to homelessness cannot eas- ily be predicted in advance, and so lend fur- ther variation to responses at the local level. While a more visible street homeless popu- lation may help induce a more punitive re- sponse to homelessness, for example, such visibility is as likely to produce an upsurge in charitable giving and in the activities of third- sector organizations. Where problems of homelessness are less apparent because less visible (whether because of urban morphology or a more extensive shelter system), other services (drop-in centers or soup kitchens, for example) may be less likely to develop.

Taken together, these four factors (path- dependent restructuring, different welfare regimes, processes of cultural signifi cation, and locally contingent variation) provide a framework through which scholars might begin to examine variations in the production of and responses to homelessness in different places, and so begin the task of developing more nuanced geographies of homelessness. Here, we illustrate this framework with some brief examples of work exploring problems of homelessness beyond the USA.

In Britain, there has been notable and r a p i d p o l i c y t r a n s f e r f r o m t h e U n i t e d States across a range of social policy arenas (Deacon, 2000; Peck, 2001), including h o m e l e s s n e s s ( C a m e r o n , 2 0 0 7 ) . A s a result, a number of towns and cities across

Britain have adopted what might be termed a ‘US-inspired’ response to the problems of on-street homelessness, including the emer- gence of Business Improvement Districts, new legislation to ban begging, and the legal exclusion of homeless people from key parts of the city centre through Anti-Social Behaviour Orders (Cameron, 2007). Dif- fering somewhat from the US experience, but cementing these developments, central government (in the form of its Homelessness Action Programme and, more recently, Single Homelessness Reviews) has also recently de- veloped a more coordinated (and more ag- gressive) response to the problems of street homelessness. This has put pressure on the local state, and upon the third-sector agen- cies operating within their jurisdiction, to ra- tion more carefully the services offered to homeless people and to move away from ‘on- street’ towards ‘off-street’ services. None- theless, and as others have argued (MacLeod, 2002; May et al., 2005; 2006), it is possible even here to recognize a more accommo- dative response to the problems of homeless- ness, with the Homelessness Action Programme also securing a significant increase in the funding available to the providers of homeless services, beyond London especially. More generally, and echoing other areas of welfare policy, homeless people in Britain enjoy a legal right of access to a greater array of welfare services than do their counterparts in the USA, including – for those with dependent children – the right to housing that is heavily (if not entirely) subsidized by the local state. As a result, the street homeless form a much smaller part of the homeless population than they do in the USA, with the vast majority of homeless people found not on the streets (or caught up in street sweeps) but in bed- and-breakfast hotels/SROs, refuges and shelters awaiting rehousing by the local state (May, 2008).

Perhaps because of borrowing more obvi- ously from UK homelessness policy rather than from the USA, but also perhaps because the morphology of New Zealand’s cities

658 Progress in Human Geography 33(5)

often renders problems of homelessness less visible than in either of these other countries, in New Zealand too a more accommodative response to homelessness is apparent. While anti-homeless policies do exist in New Zealand (with a number of cities targeting an increase in levels of begging in recent years, for example) local authorities have tended to respond to rising levels of homelessness with funding for supported housing and social welfare programs rather than with anti- homeless ordinances (May, 2004; Laurenson and Collins, 2007).

As our last example makes clear, local variations in the cultural signifi cation of home- lessness also make a difference to the ways in which problems of homelessness are re- sponded. Thus, as Song (2006) has shown, although different countries may construct apparently similar distinctions between the ‘deserving’ and the ‘undeserving’ homeless, and target their responses to different home- less groups accordingly, the lines of demar- cation themselves may be drawn rather differently. In Seoul, for example, these dis- tinctions are drawn not between single homeless people and homeless families (as in the UK, or the USA) but between the long- term and more recently homeless. While the former have been subject to increased sur- veillance, clearance and control in recent years, the latter have benefi ted from more accommodative responses, including the provision of new shelters designed to help ‘reintegrate’ homeless men back in to main- stream housing and employment.

As these examples make clear, neither the problems of, nor the response to, homeless- ness are the same in Britain, New Zealand and South Korea as they are in the United States. In light of such differences it is imperative that geographers interested in homelessness broaden their focus beyond the USA to build a more nuanced picture of the geographies of homelessness. Indeed, they would do well to look beyond the Minority World. The scale of both homelessness and the aggressive, often deadly, attempts to

deal with it takes on a new dimension in cities such as São Paulo, Cairo, Mumbai or Manilla, where upwards of half the population live in precarious or informal shelter and where tens of thousands of children live (and die) on the streets (Dos Santos, 2003; Roy, 2003; Davis, 2006; Speak and Tipple, 2006). Geog- raphers working on homelessness have been remarkably reluctant to engage with the problems of homelessness in the Majority World (though see Beazley, 2002), even though it is plain that such an engagement would signifi cantly enrich the geographical imaginations of those working in both urban and development studies (Robinson, 2002).

VI Differentiation, difference, identity … and agency As Takahashi (1996) made clear, a fi nal fac- tor shaping responses to homelessness is variation within the homeless population itself. One of the most serious limitations of work proceeding within the punitive frame is a tendency to refer to homeless people as a homogenous and largely androgynous group: ‘The homeless’. Such reductionism is all the more surprising given wider sensitivities to difference within the discipline and the grow- ing awareness of the need to differentiate between different publics (Ward, 2006).

In the current context, it is now clear that race and gender, for example, shape the ex- perience of and responses to homelessness in complex ways. Thus, while some have begun to explore the greater sympathy ac- corded to homeless women with children, for example (Baker, 1994; Salomon et al., 1996; Takahashi et al., 2002), others have recognized the increased stigma faced by single ‘unaccommodated women’ who often find themselves ill catered for by services which still tend to target homeless men (DeVerteuil, 2003; May et al., 2007). Re- search on the engendering of homelessness is still in its infancy, with geographers yet to contribute to discussions around the inter- sections between homelessness and mascu- linity; for example, see Higate (2000a; 2000b).

Geoffrey DeVerteuil et al.: Complexity not collapse 659

But it is at least now (back) on the agenda (see also Takahashi, 1998; DeVerteuil, 2005; Whitzman, 2006), with recent calls for more attention to be paid by geographers to the role of gender in shaping numerous as- pects of homelessness, whether homeless people’s experiences on the streets, access to services, or indeed constructions of ‘home’ and ‘homelessness’ themselves (Klodawsky, 2006; for much earlier research on the en- gendering of homelessness, see Rowe and Wolch, 1990).

Much less has been said, by geographers at least, about the intersections of homeless- ness and race, or homelessness, race and gen- der (for a review of work on homelessness and race beyond geography, see Harrison, 1999). As Passaro (1996) has argued, when it comes to race the impression too often is that, al- though the chances of becoming homeless may differ signifi cantly by race, race matters little once an individual is homeless. As Passaro suggests:

The people who are persistently homeless in Manhattan are, overwhelmingly, black men. Why? The answer, I found, is that homelessness is not simply an economic predicament, but a cultural and moral location … [Homeless] men appear to be independent of the control of women, family, and society, and thus they are considered dangerous, violent, and aggressive. If they are ‘non-white’, racism reinforces and exaggerates these fears. (Passaro, 1996: 1)

Rather than examining how questions of race shape constructions of homelessness, or how race and ethnicity shape people’s experiences of homelessness, geographers working in this fi eld have mainly focused on issues of environmental racism: examining why it is that homeless facilities are more likely to be located in minority ethnic areas than in white (suburban) neighborhoods (see, for example, Hopper and Baumohl, 1994; Wright, 1997). But scholars have also begun to examine the ‘locational preferences’ of homeless people themselves, differentiating these by ethnicity. In a study of homeless

women’s movements across the Los Angeles shelter system, for example, DeVerteuil (2001; 2003) found a clear pattern of avoidance by white homeless women of heavily minority areas, particularly Downtown where the vast majority of homeless people are African- American. By contrast, African-American women were more likely to stay near South-Central Los Angeles, the traditional heartland of Los Angeles’s African-American community and an area where many of the women came from. The few immigrant women included in the study (the majority of whom were Latina) similarly stuck to their community areas and support networks, and studiously avoided mainstream homeless services such as the Salvation Army (see also Molina, 2000).

This latter work is especially important, perhaps, because with its focus on ‘prefer- ence’ (however limited the options within which such preferences operate) it returns a sense of agency to homeless people. Too often in work framed by the punitive turn it appears as though homeless people lack any agency at all: the hapless victims of mass evictions and street sweeps. This is not to say that such accounts entirely ignore the pos- sibility of resistance to recent anti-homeless measures. Rather, when resistance comes it is ‘Resistance’ with a capital R: riots, mass occupations (Smith, 1992) or legal challenges to anti-homeless legislation (Mitchell, 2003). Such moments are clearly important, but they are also quite rare. Furthermore, these acts of resistance are very often organized not by homeless people but by activists working on their behalf (but see Dear and von Mahs, 1997; Wright, 1997).

Indeed, while we frequently hear the views of those seeking to banish homeless people from the city, homeless people themselves are rarely given voice in such accounts – even in mediated form. As a result we learn next to nothing about their circumstances: how and why they became homeless, for example, how long they have been homeless, their experiences of homeless services or of

660 Progress in Human Geography 33(5)

the wider city – or whether even they are black or white, young or old, male or female (Mohan, 2002). Although such work is clearly framed by an oppositional politics, one could be forgiven for suggesting that the interest of those writing within the punitive frame lies less with the geographies of homeless- ness per se, or with homeless people, than with using homeless people as convenient ciphers around which to build a wider cri- tique of gentrifi cation, public space law and so on: ‘homelessness … [as the] bellwether of urban justice’, as Mitchell himself has put it (2003: 9).

If such is the case, the question we must presumably ask is: does such a tendency matter? We think it does, for at least three reasons. First, experiences of homelessness – including whether or not a person is liable to find themselves subject to the kind of regressive policies these authors outline – differ considerably across a heterogeneous homeless population. If the real interest of these authors is in exploring the changing politics of public space, then we need to be aware that the effect of such changes are liable to be experienced differently by different homeless people. Second, it matters because such differences – how long a person has been homeless, their age, race or gender, for example – are crucial to developing better public policy responses to the problems of homelessness, whether affordable housing programs, tenancy support packages, rehab programs, or a more diverse array of emer- gency shelters designed to serve an equally diverse homeless population. Third, in their failure to recognize differences within the homeless population, to give voice to home- less people themselves, or recognize the (many and diverse) ways in which homeless people continue to ‘get by’ and survive in the city (Ruddick, 1996), such accounts replicate the very move for which they so vigorously criticize others – ‘denying homeless people … [their] individuality … humanity’ and agency (Mitchell, 2003: footnote 21, 158; see also his defense of exactly this tendency in his

own work – footnote 21, 158–59). Such a tendency is all the more frustrating

given the rich vein of work, a good deal of it by geographers, on which those interested in exploring homeless people’s experiences of the city could now draw. Such work deploys a range of often innovative methodologies including ethnography (Cloke et al., 2008), participant observation (Cooper, 2001), life history approaches (May, 2000; von Mahs, 2005), and autophotography (Johnsen et al., 2008) in order to give narrative space to homeless people’s own accounts of city life. Drawing on a biographical method, for example, von Mahs (2005) has examined the experiences of single homeless people in Germany, differentiating between respond- ents on the basis of their housing and home- less histories. He shows that people with more ‘regular’ life courses are less likely to experience the kind of harassment traced in the punitive literature than those with more ‘irregular’ life courses (a history of residential instability and previous homeless episodes). Yet when asked to refl ect on their experiences of regulation and policing, even this latter group reported that their main concerns related not to this harassment, but to the problems experienced in accessing affordable housing and employment.

Others have looked in detail at the ways in which homeless people’s movements around the city are circumscribed by the dif- ferent values ascribed to urban space by the domiciled population (whether ‘prime’ or more ‘marginal’ spaces, for example – a con- tinuum we fi nd more useful than the simple division between ‘public’ and ‘private’ space) (Ruddick, 1996; Wardhaugh, 2000); at the ways in which homeless people attempt to reinscribe spaces of ‘home’ across the wider cityscape (Veness, 1993; Ruddick, 1996; Cooper, 2001); and at the many and varied ways in which people ‘get by’ on the streets (whether fi nding a safe place to sleep, earning money, drawing upon networks of mutual support, or dealing with more inten- sive policing and surveillance) (Snow and

Geoffrey DeVerteuil et al.: Complexity not collapse 661

Anderson, 1993; Lees, 1997; Knowles, 2000). Rather than stressing dramatic moments of Resistance, or the strategies devised by ac- tivists to challenge recent anti-homeless initiatives, such work draws attention to the many, varied and creative tactics homeless people themselves draw upon in order to carve out a space in the city (for an account that moves beyond this ‘rational’ reading of the homeless city and stresses instead the emotional and performative aspects of home- lessness, see Cloke et al., 2008).

Instead of victims, homeless people emerge in these accounts as highly tenacious, showing a remarkable ability to cope in the face of attempts to ‘sweep’ them from the streets – as Duneier observes:

Many politicians incorrectly think that, if only they were to pass more laws to make life diffi cult for disreputable people working the sidewalks, these people would go away … [But] surely, the people working and/or living on the sidewalks cannot be done away with so easily, even if it were desirable. Those de- termined to make ‘an honest living’ will keep deploying their creativity, competence, and cultural knowledge, as the men and women on Sixth Avenue do, to survive. (Duneier, 1999: 312)

VII Conclusions Over the past decade or so there has been a marked proliferation of work by geographers on homelessness. Inspired by the writings of a few key authors, much of this work has been framed by the desire to connect the question of homelessness to wider debates around gentrifi cation, urban restructuring and the politics of public space. Such work has been enormously useful, shifting discussions of homelessness in to the mainstream geograph- ical literature.

While welcoming the growing interest in problems of homelessness, we share the con- cerns recently expressed by others about such work (Laurenson and Collins, 2007; Murphy, 2009): namely, that too much of it remains narrowly framed within a US metric of knowledge, and too closely focused upon

the recent punitive turn in urban social policy. While it is clear to us that the past decade has seen a range of increasingly punitive measures bear down upon homeless people in a growing number of cities across north America, western Europe and beyond, it is equally clear that recent responses to homelessness have not been ‘uniformly hostile’ (Laurenson and Collins, 2007: 650). Rather, just as we can trace significant differences in the extent and form that homelessness takes in different places, so too we can trace a range of responses to the problems of homelessness: from the obvi- ously punitive, to the ambivalent, to the more accommodating. Far from ‘collapsing’, the geographies of homelessness have become more varied and more complex in recent years, as geographically variable responses to the problems of homelessness have led to a recent proliferation, and disarticulation, of spaces of homelessness across the contem- porary western city.

In other words, there is a complex and varied geography of homelessness and of re- sponses to homelessness: a geography that has too often been ignored in recent writings on homelessness by geographers. Moving beyond a critique of the ‘punitive turn’, we have therefore outlined a means of conceptu- alizing these more varied responses – poverty management – and advanced a new frame- work through which we might begin to develop a better understanding of this geo- graphical variation in both the production of and responses to homelessness.

But we would also argue that if the interest really is in the geographies of homelessness and the needs of homeless people – rather than in the politics of public space or of the ‘revanchist/post-justice’ city – then scholars also need to engage more directly with the real authors of these homeless geographies: city managers, welfare offi cials, voluntary- sector organizations and homeless people themselves. Only when we construct ac- counts of the homeless city that these authors recognize might we begin to shape these

662 Progress in Human Geography 33(5)

geographies. This means constructing ac- counts that draw attention to the regressive policies unfolding in many cities, certainly, but which also give: space to the battles raging between city managers, public policy offi cials and welfare agencies over the most appro- priate response to homelessness; credence to the attempts of the thousands of voluntary- sector organizations, and volunteers, strug- gling to respond to the needs of homeless people; and voice to homeless people them- selves so we can better understand their experiences of homelessness, and champion their hopes and aspirations. Far from hapless, we might then fi nd that recent developments give grounds for hope: hope in the urge to care for homeless people that is still every bit as apparent, and powerful, as a ‘revanchist’ mindset; and hope that geographers may actually begin (as some have already begun) to formulate their own suggestions as to how best to tackle the problems of homelessness, rather than to continue only to decry them.

References Alcock, P. and Craig, G., editors 2001: International

social policy: welfare regimes in the developed world. Basingstoke: Palgrave.

Arapoglou, V.P. 2004: The governance of homeless- ness in the European South: spatial and institutional contexts of philanthropy in Athens. Urban Studies 41, 621–39.

Baker, S. 1994: Gender, ethnicity, and homelessness. American Behavioral Scientist 37, 476–504.

Beazley, H. 2002: ‘Vagrants wearing make-up’: nego- tiating spaces of the streets of Yogyakarta, Indonesia. Urban Studies 39, 1665–83.

Blomley, N. 2006: Editorial: Homelessness and the delusions of property. Transactions of the Institute of British Geographers NS 31, 3–5.

— 2007a: How to turn a beggar into a bus stop: law, traffi c and the ‘function of the place’. Urban Studies 44, 1697–712.

— 2007b: Legal geographies – Kelo, contradiction, and capitalism. Urban Geography 28, 198–205.

Blomley, N. and Klodawsky, F. 2007a: Rights, space and homelessness 1. Session convened at the annual conference of the Association of American Geo- graphers, San Francisco, April.

— 2007b: Rights, space and homelessness 2. Session convened at the annual conference of the Association of American Geographers, San Francisco, April.

— 2007c: Rights, space and homelessness 3. Session convened at the annual conference of the Association of American Geographers, San Francisco, April.

Brenner, N. 2004a: Urban governance and the pro- duction of new state spaces in western Europe, 1960–2000. Review of International Political Economy 11, 447–88.

— 2004b: Stereotypes, archetypes and prototypes: three uses of superlatives in contemporary urban studies. City and Community 3, 205–18.

Brinegar, S. 2003: The social construction of homeless shelters in the Phoenix area. Urban Geography 24, 61–74.

Burt, M., Aron, L,. Lee, E. and Valentine, J. 2001: Helping America’s homeless: emergency shelter or affordable housing. Washington, DC: The Urban Institute Press.

Cameron, A. 2007: Geographies of welfare and ex- clusion: reconstituting the ‘public’. Progress in Human Geography 31, 519–26.

Cloke, P., Johnsen, S. and May, J. 2005: Exploring ethos? Discourses of charity in the provision of emergency services for homeless people. Environment and Planning A 37, 385–402.

— 2007a: Ethical citizenship? Volunteers and the ethics of providing services for homeless people. Geoforum 38, 1089–101.

— 2007b: The periphery of care: emergency services for homeless people in rural areas. Journal of Rural Studies 23, 387–401.

Cloke, P. May, J. and Johnsen, S. 2008: Performativity and affect in the homeless city. Environment and Planning: Society and Space 26, 241–63.

Cloke, P., Milbourne, P. and Widdowfield, R. 2000: Homelessness and rurality: ‘out of place’ in purifi ed space? Environment and Planning D: Society and Space 18, 715–35.

Coleman, R. 2004: Images from a neoliberal city: the state, surveillance and social control. Criminology 12, 21–42.

Collins, D. and Blomley, N. 2003: Private needs and public space: politics, poverty, and anti-panhandling by-laws in Canadian cities. In Law Commission of Canada, editor, New perspectives on the public– private divide, Vancouver: University of British Columbia, 40–68.

Cooper, R. 2001: The intersection of space and home- lessness in Central Auckland. Unpublished MA thesis, Department of Geography, University of Auckland, NZ.

Cowan, D. 2006: Adjudicating the implementation of homeless law: the promise of socio-legal studies. Housing Studies 21, 381–400.

Cresswell, T. 1996: In place/out of place: Geography, ideology and transgression. Minneapolis, MN: University of Minnesota Press.

Daly, M. 1999: Regimes of social policy in Europe and the patterning of homelessness. In Avramov, D.,

Geoffrey DeVerteuil et al.: Complexity not collapse 663

editor, Coping with homelessness: issues to be tackled and best practices in Europe, Aldershot: Ashgate, 309–30.

Dangschat, J. 1997: Local political reactions on the surplus population of Hamburg. Paper presented at Homelessness and Urban Restructuring: the Americanization of the European City? The Contradictory Geography of Socio-spatial Injustices in Global(izing) Cities, John F. Kennedy Institute, University of Berlin, October.

Datta, A. 2005: ‘Homed’ in Arizona: the architecture of emergency shelters. Urban Geography 26, 536–57.

Davis, M. 1990: City of quartz. London: Vintage. — 2006: Planet of slums. London: Verso. Deacon, A. 2000: Learning from the US? The infl uence

of American ideas upon ‘new labour’ thinking on welfare reform. Policy and Politics 28, 5–18.

Dean, H., editor 1999: Begging questions: street level economic activity and social policy. London: Policy Press.

Dear, M. and von Mahs, J. 1997: Housing for the homeless, by the homeless, and of the homeless. In Ellin, N., editor, The architecture of fear, Princeton, NJ: Princeton University Press, 187–200.

DelCasino, V. and Jocoy, C. 2008: Neoliberal sub- jectivities, the ‘new’ homelessness, and struggles over spaces of/in the city. Antipode 40, 192–99.

DeVerteuil, G. 2001: Welfare reform and welfare neighborhoods: institutional and individual perspec- tives. Unpublished PhD thesis, Department of Geography, University of Southern California.

— 2003: Homeless mobility, institutional settings, and the new poverty management. Environment and Planning A 35, 361–79.

— 2004: Systematic inquiry into barriers to researcher access: evidence from a homeless shelter. The Pro- fessional Geographer 56, 372–80.

— 2005: The relationship between government assist- ance and housing outcomes among extremely low-income individuals: a qualitative inquiry in Los Angeles. Housing Studies 20, 383–99.

— 2006: The local state and homeless shelters: beyond revanchism? Cities 23, 109–20.

Dordick, G. 1997: Something left to lose: personal re- lations and survival among New York’s homeless. Philadelphia, PA: Temple University Press.

Dos Santos, C. 2003: Cities of plastic and cardboard: the informal habitat of homeless people in Sao Paulo, Los Angeles and Tokyo. Professorship thesis, University of São Paulo.

Duneier, M. 1999: Sidewalk. New York: Farrar, Straus and Giroux.

Esping-Andersen, G. 1990: The three worlds of welfare capitalism. Cambridge: Polity Press.

Evans, L. 2007: Housing justice welcomes London Councils’ decision to drop soup run ban. Retrieved 25

March 2009 from http://www.housingjustice.org. uk/forum/viewtopic.php?p=56&sid=a44162dc71217 56b56fcf869f03c549a

Fitzpatrick, S. 2001: The links between begging and rough sleeping: a question of legitimacy? Housing Studies 16, 549–68.

Fopp, R. 2002: Increasing the potential for gaze, sur- veillance and normalization: the transformation of an Australian policy for people who are homeless. Surveillance and Society 1, 48–65.

Gaber, S.L. 1996: From NIMBY to Fair Share: the de-velopment of New York City’s municipal shelter siting policies, 1980–1990. Urban Geography 17, 294–316.

Glasser, I. 1994: Homelessness in global perspective. Boston, MA: G.K. Hall and Company.

Harrison, M. 1999: Theorizing homelessness and ‘race.’ In Kennett, P. and Marsh, A., editors, Home- lessness: exploring the new terrain, Bristol: Policy Press, 101–20.

Harvey, D. 2005: A brief history of neoliberalism. Oxford: Oxford University Press.

Herbert, S. and Brown, E. 2006: Conceptions of space and crime in the punitive neoliberal city. Antipode 38, 4, 755–77.

Hetherington, K. 2001: Moderns as ancients: time, space and the discourse of Improvement. In May, J. and Thrift, N., editors, TimeSpace: geographies of temporality, London: Routledge, 49–72.

Higate, P. 2000a: Ex-servicemen on the road: travel and homelessness. The Sociological Review 48, 331–48.

— 2000b: Tough bodies and rough sleeping: embodying homelessness among ex-servicemen. Housing, Theory and Society 17, 97–108.

Hoch, C. 1991: The spatial organization of the urban homeless: a case study of Chicago. Urban Geography 12, 137–54.

Hopper, K. and Baumohl, J. 1994: Held in abeyance: rethinking homelessness and advocacy. American Behavioral Scientist 37, 522–52.

Hubbard, P. 2004: Revenge and injustice in the neoliberal city: uncovering masculinist agendas. Antipode 36, 665–86.

Huber, E. and Stephens, J. 2001: Development and crisis of the welfare state: parties and policies in global markets. Chicago: Chicago University Press.

Inter-University Consortium Against Homelessness 2007: A reality-based approach to ending homelessness in Los Angeles. Los Angeles: Inter-University Consortium Against Homelessness.

Johnsen, S. and Fitzpatrick, S. 2007: The impact of enforcement on street users in England. Plymouth: The Policy Press.

Johnsen, S., Cloke, P. and May, J. 2005a: Day cen- tres for homeless people: spaces of care or fear? Social and Cultural Geography 6, 787–811.

664 Progress in Human Geography 33(5)

— 2005b: Transitory spaces of care: serving homeless people on the street. Health and Place 11, 323–36.

Johnsen, S., May, J. and Cloke, P. 2008: Imag(in)ing h o m e l e s s p l a c e s : u s i n g a u t o - p h o t o g r a p h y t o (re)examine the geographies of homelessness. Area 40, 194–207.

Katz, C. 2001: Hiding the target: social reproduction in the privatized urban environment. In Minca, C., editor, Postmodern geography: theory and praxis, Oxford: Blackwell, 93–110.

Klodawsky, F. 2006: Landscapes on the margins: gender and homelessness. Gender, Place and Culture 13, 365–81.

Knowles, C. 2000: Bedlam on the streets. London: Routledge.

Larner, W. 2000: Neoliberalism: policy, ideology, gov- ernmentality. Studies in Political Economy 63, 5–26.

Laurenson, P. and Collins, D. 2006: Towards inclusion: local government, public space and homelessness in New Zealand. New Zealand Geographer 62, 185–95.

— 2007: Beyond punitive regulation? New Zealand local governments’ responses to homelessness. Antipode 39, 649–67.

Law, R. 2001: ‘Not in my city’: local governments and homelessness policies in the Los Angeles Metropolitan Region. Environment and Planning C 19, 791–815.

Laws, G. 1992: Emergency shelter networks in an urban area: serving the homeless in Metropolitan Toronto. Urban Geography 13, 99–126.

Lee, B. and Farrell, C. 2003: Buddy, can you spare a dime? Homelessness, panhandling, and the public. Urban Affairs Review 38, 299–324.

Lees, L. 1997: Ageographia, heterotopia and Vancouver’s new public library. Environment and Planning D: Society and Space 15, 321–47.

— 2005: The ambivalence of diversity and the politics of urban renaissance: the case of youth in downtown Portland, Maine. International Journal of Urban and Regional Research 27, 613–34.

Leitner, H. 1992: Urban geography: responding to new challenges. Progress in Human Geography 16, 105–18.

Link, B.G., Schwartz, S., Moore, R., Phelan, J., Struening, E., Stueve, A. and Colten, M.E. 1995: Public knowledge, attitudes, and beliefs about homeless people: evidence for compassion fatigue? American Journal of Community Psychology 23, 533–55.

Lobao, E. and Murray, A. 2005: Exploratory analysis of the homeless shelter system in Columbus, Ohio. Geografi ska Annaler B 87, 1, 61–73.

MacLeod, G. 2002: From urban entrepreneurialism to a ‘revanchist city’? On the spatial injustices of Glasgow’s renaissance. Antipode 34, 602–24.

Mair, A. 1986: The homeless and the post-industrial city. Political Geography Quarterly 5, 351–68.

Marr, M. 1997: Maintaining autonomy: the plight of the American Skid Row and Japanese Yoseba. Journal of Social Distress and the Homeless 6, 229–50.

Mathieu, A. 1993: The medicalization of homelessness and the theatre of repression. Medical Anthropology Quarterly 7, 170–84.

May, J. 2000: Of nomads and vagrants: single homeless- ness and narratives of home as place. Environment and Planning D: Society and Space 18, 737–59.

— 2003: The view from the streets: geographies of homelessness in the British newspaper press. In Blunt, A., Grufudd, P., May, J., Ogborn, M. and Pinder, D. editors, Practising cultural geography, London: Arnold, 23–38.

— 2004: 12 things wrong with the revanchist city thesis. In Beyond the Revanchist City, Annual Meeting of the Association of American Geographers, Philadelphia, March.

— 2008: Homelessness. In Kitchin, R. and Thrift, N., editors, International encyclopaedia of human geography, London: Elsevier.

May, J., Johnsen, S. and Cloke, P. 2005: Re-phasing neo-liberalism: New Labour and Britain’s crisis of street homelessness. Antipode 37, 703–30.

— 2006: Shelter at the margins: New Labour and the changing state of emergency accommodation for single homeless people in Britain. Policy and Politics 34, 711–30.

— 2007: Alternative cartographies of homelessness: rendering visible British women’s experiences of ‘visible’ homelessness. Gender, Place and Culture 14, 121–40.

Merrifi eld, A. 2000: The dialectics of dystopia: disorder and zero tolerance in the city. International Journal of Urban and Regional Research 24, 473–89.

Metraux, S. 1999: Waiting for the wrecking ball: Skid Row in postindustrial Philadelphia. Journal of Urban History 25, 690–715.

Milligan, C. and Fyfe, N. 2004: Putting the voluntary sector in its place: geographical perspectives on voluntary activity and social welfare in Glasgow. Journal of Social Policy 33, 73–93.

Minnery, J. and Greenhalgh, E. 2007: Approaches to Homelessness Policy in Europe, the United States, and Australia. Journal of Social Issues 63, 641–55.

Mitchell, D. 1997: The annihilation of space by law: the roots and implications of anti-homeless laws in the United States. Antipode 29, 303–36.

— 1998a: Anti-homeless laws and public space I: Begging and the First Amendment. Urban Geography 19, 6–11.

— 1998b: Anti-homeless laws and public space II: further constitutional issues. Urban Geography 19, 98–104.

— 2001: Postmodern geographical praxis? The post- modern impulse and the war against homeless people in the ‘post-justice’ city. In Minca, C., editor, Postmodern geography: theory and praxis, Oxford: Blackwell, 57–92.

— 2003: The right to the city: social justice and the fi ght for public space. London: Guilford.

Geoffrey DeVerteuil et al.: Complexity not collapse 665

— 2005: The S.U.V. model of citizenship, floating bubbles, buffer zones, and the rise of the ‘purely atomic’ individual. Political Geography 24, 77–100.

Mohan, J. 2002: Geographies of welfare and social exclusion: dimensions, consequences and methods. Progress in Human Geography 26, 65–75.

Molina, E. 2000: Informal non-kin networks among homeless Latino and African American men: form and functions. American Behavioral Scientist 43, 663–85.

Murphy, S. 2009: ‘Compassionate’ strategies of man- aging homelessness: post-revanchist geographies in San Francisco. Antipode 41, 305–25.

Neale, J. 2001: Homelessness among drug users: a double jeopardy explored. International Journal of Drug Policy 12, 353–69.

Passaro, J. 1996: The unequal homeless: men on the street, women in their place. New York: Routledge.

Peck, J. 2001: Neo-liberalizing states: thin policies/hard edges. Progress in Human Geography 25, 445–55.

Piven, F.F. and Cloward, R. 1993: Regulating the poor: the functions of public welfare (second edition). New York: Vintage Books.

Robinson, J. 2002: Global and world cities: a view from off the map. International Journal of Urban and Regional Research 26, 531–54.

Rowe, S. and Wolch, J. 1990: Social networks in time and space: homeless women in Skid Row, Los Angeles. Annals of the Association of American Geographers, 80, 184–205.

Roy, A. 2003: City requiem, Calcutta: gender and the politics of poverty. Minneapolis, MN: University of Minnesota Press.

Ruddick, S. 1996: Young and homeless in Hollywood: mapping social identities. New York: Routledge.

Salomon, A., Bassuk, S. and Brooks, M. 1996: Patterns of welfare use among poor and homeless women. American Journal of Orthopsychiatry 66, 510–25.

Slater, T. 2004: North American gentrification? Revanchist and emancipatory perspectives explored. Environment and Planning A 36, 1191–213.

Smith, N. 1992: New city, new frontier: the Lower East Side as Wild West. In Sorkin, M., editor, Variations on a theme park: the new American city and the end of public space, New York: Hill and Wang, 61–93.

— 1996a: The new urban frontier: gentrifi cation and the revanchist city. London: Routledge.

— 1996b: Social justice and the new American urbanism: the revanchist city. In Merrifi eld, A. and Swyngedouw, E., editors, The urbanization of in- justice, London: Lawrence and Wishart, 117–36.

— 1998: Giuliani time: the revanchist 1990s. Social Texts 57(18), 1–20.

— 2001: Global social cleansing: postliberal revanchism and the export of zero tolerance. Social Justice 28(3), 68–74.

Snow, D. and Anderson, L. 1993: Down on their luck: a study of homeless street people. Los Angeles, CA: University of California Press.

Snow, D. and Mulcahy, M. 2001. Space, politics, and the survival strategies of the homeless. American Behavioral Scientist 45, 149–69.

Song, J. 2006: Historicization of homeless spaces: the Seoul train station square and the house of freedom. Anthropological Quarterly 79, 193–223.

Sorkin, M., editor 1992: Variations of a theme park: the new American city and the end of public space. New York: Hill and Wang.

Speak, S. and Tipple, G. 2006: Perceptions, persecu- tion and pity: the limitations of interventions for homelessness in developing countries. International Journal of Urban and Regional Research 30, 172–88.

Staeheli, L. and Mitchell, D. 2007: Locating the public in research and practice. Progress in Human Geography 31, 792–811.

Swanson, K. 2007: Revanchist urbanism heads south: the regulation of indigenous beggars and street vendors in Ecuador. Antipode 39, 708–28.

Takahashi, L. 1996: A decade of understanding homelessness in the USA: from characterization to representation. Progress in Human Geography 20, 291–31.

— 1997: The socio-spatial stigmatization of home- lessness and HIV/AIDS: toward an explanation of the NIMBY syndrome. Social Science and Medicine 45, 903–14.

— 1998: Homelessness, AIDS, and stigmatization: the NIMBY syndrome in the United States at the end of the twentieth century. Oxford: Clarendon Press.

Takahashi, L., McElroy, J. and Rowe, S. 2002: The socio-spatial stigmatization of homeless women with children. Urban Geography 23, 3102–22.

Uitermark, J. and Duyvendak, J. 2008: Civilizing the city: populism and revanchist urbanism in Rotterdam. Urban Studies 45, 1485–503.

Veness, A.R. 1993: Neither homed nor homeless: contested defi nitions and the personal worlds of the poor. Political Geography 12, 319–40.

Veness, I. 1994: Designer shelters as models and makers of home: new responses to homelessness in urban America. Urban Geography 15, 150–67.

von Mahs, J. 2005: The socio-spatial exclusion of homeless people in Germany and the United States. Special issue on homelessness and social exclusion. American Behavioral Scientist 48, 928–60.

Wacquant, L. 1999: Urban marginality in the coming millennium. Urban Studies 36, 1639–47.

Ward, K. 2006: Public intellectuals, geography, its re- presentations, and its publics. Geoforum 38, 1058–64.

Wardhaugh, J. 2000: Sub City: young people, homeless- ness and crime. London: Ashgate.

666 Progress in Human Geography 33(5)

Whitzman, C. 2006: At the intersections of invisi- bilities: Canadian women, homelessness and health outside the ‘big city’. Gender, Place and Culture 13, 185–95.

Williams, J.C. 1996: Geography of the homeless shelter: staff surveillance and resident resistance. Urban Anthropology 25, 75–113.

Wolch, J. 1996: From global to local: the rise of home- lessness in Los Angeles during the 1980s. In Scott, A. and Soja, E., editors, The city, Berkeley, CA: University of California Press, 390–25.

— 1997: Homelessness and welfare reform. Paper pres- ented at Homelessness and Urban Restructuring: the Americanization of the European City? The Contradictory Geography of Socio-spatial Injustices in Global(izing) Cities, John F. Kennedy Institute, University of Berlin, October.

— 2008: Intransigent LA. Geoforum 39, 543–45. Wolch, J. and Dear, M. 1993: Landscapes of despair:

from deinstitutionalization to homelessness. Princeton, NJ: Princeton University Press.

Wolch, J. and DeVerteuil, G. 2001: Landscapes of the new poverty management. In May, J. and Thrift, N., editors, TimeSpace, London: Routledge, 149–68.

Wright, T. 1997: Out of place: homeless mobilizations, subcities, and contested landscapes. Albany, NY: State University of New York Press.

Wyly, E. and Hammel, D. 2005: Mapping neoliberal American urbanism. In Atkinson, R. and Bridge, G., editors, Gentrifi cation in a global context: the new urban colonialsim, London: Routledge, 18–38.

Zukin, S. 1995: The cultures of cities. New York: Blackwell.