Homelessness and Community Author(S): Jeremy Waldron Source: the University of Toronto Law Journal, Vol

Home , Jeremy Waldron

Homelessness and Community Author(s): Jeremy Waldron Source: The University of Toronto Law Journal, Vol. 50, No. 4 (Autumn, 2000), pp. 371-406 Published by: University of Toronto Press Stable URL: http://www.jstor.org/stable/825960 . Accessed: 06/01/2014 13:51

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

University of Toronto Press is collaborating with JSTOR to digitize, preserve and extend access to The University of Toronto Law Journal.

http://www.jstor.org

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions JeremyWaldron* HOMELESSNESS AND COMMUNITYt

I Introduction

I am not a communitarian,but I was prompted to ask questions about the relation between homelessness and communityby an encounter in Californiasome yearsago withan organizationcalled The AmericanAlli- ance forRights and Responsibilities(AARR). The AARRis an activistarm of The CommunitarianNetwork,' litigating to defend local and municipal initiativesthat promote what it regardsas increased safety,civility, and

* Mauriceand Hilda FriedmanProfessor of Law and Directorof theCenter for Law and Philosophy,Columbia University. t This is a slightlymodified version of the Cecil A. WrightMemorial Lecture, deliveredat theUniversity of Toronto Law School in February2000. Earlier versions were presented, under the title Homelessness,Community and Denial, at a 1998 Con- ferenceon Homelessnessat GeorgiaState University in Atlantaand at a meetingof Columbia University'sFifteen Minute Paper Group. I am gratefulto all the participantson thoseoccasions - particularlyRobert Ellickson - fortheir comments and suggestions. 1 D.B. Kopel & C.C. Little,'Communitarians, Neorepublicans, and Guns:Assessing the Case For Firearms Prohibition'(1997) 56 Mar.L.Rev438 at 443: 'The CommunitarianNetwork has createdan activistarm to implementits ideas on a grassrootslevel: the American Alliance for Rights and Responsibilities.'The AARRis a non-profitpublic interestgroup founded in 1988 and describedby one of its leading litigatorsas aimingto 'restorethe spiritof communityin the United States....[Its] approachis to identify,promote and defendnew approaches which make citizenspart of the solution,and strikea balance betweenextreme rights claimsand thosewho would sacrifice civil liberties as meansto an end.' R.S.Golden, 'Towarda Model of CommunityRepresentation for Legal AssistanceLawyering: Examiningthe Role of Legal AssistanceAgencies in Drug-relatedEvictions from PublicHousing' (1988) 17 Yale L.& Pol'y Rev. 527 at 552n,quoting Affidavit of RogerConner in Supportof Motion To Modifyand Intervene,in Escalerav. NYCHA, 924 F. Supp. 1323 (S.D.N.Y.1996). In recentyears, the AARR has takenon a new name, 'The Centerfor the CommunityInterest'; see H.P. Fahringer,'Zoning Out Free Expression:An Analysis of NewYork City's Adult Zoning Resolution' (1998) 46 Buff.L.Rev.403 at 403n.The followingarticles by Robert Teir, leading counsel for AARR,provide a good accountof the organization'smission: 'Restoring Order in UrbanPublic Spaces' (1998) 2 Tex.Rev.L.& Pol. 255; 'MaintainingSafety and Civility in Public Spaces: A ConstitutionalApproach to AggressiveBegging' (1993) 54 Lou.L.Rev.285 [hereinafter'Maintaining Safety and Civility].

(2000), 50 UNIVERSITYOF TORONTO LAWJOURNAL

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions 372 UNIVERSITYOF TORONTO LAWJOURNAL communityresponsibility.2 The AARRfiled an amicus briefsupporting the Cityof San Francisco in some litigationbrought by homeless advocates against formerMayor FrankJordan's 'Matrix' program,3which was intended as a coordinated set of initiativesby police and social servicesto remove encampments of homeless people from the centre of San Francisco (particularlythe public places around the Civic Center).4 An ex-student of mine was involved in the litigation,representing the plaintiffs,and, since he had read an article on homelessnessthat I published in 1991,5he asked if I would be willingto file a response to the AARRbrief. Needless to say,the lawsuitwas unsuccessful;actually it was eventuallydeclared moot, the Cityof San Francisco having abandoned the program in question afterthe election of Mayor Willie Brown in 1995. But the issues that were raised in the exchange I had with the AmericanAlliance forRights and Responsibilitiesremained withme, and I want to talkabout some of those in thisessay.

II Communitycontrol of public places

The line taken by the AARRwas a familiarone, and it is one theyhave pursued in a number of cities in the United States.6Mayor Jordan's

2 Caseswith AARRintervention have included Asquithv. City ofBeaufort, 139 F.3d (1998), supportingSouth Carolina municipal noise ordinanceagainst challenge by street preachers;Johnson v. Rodriguez,110 F.3d 299 (1997), supportingconsideration by Texas paroleboards of letters from members of public against prisoner release; Doe v.Pataki, 120 F.3d 1263(1997), defending 'Megan's Law' requirementof registration ofsex offenders;Hutchins v. Districtof Columbia, 942 F.Supp.665 (1996), defending JuvenileCurfew Act; Herndonv. ChapelHill-Carrboro City Board ofEducation, 89 F.3d 174 (1996), defendingcompulsory public serviceprogram for high school students; Roulettev. Cityof Seattle, 78 F.3d 1425 (1996), supportingcity laws against obstruction ofsidewalk; Montana v. Egelhoff518 U.S. 37, 116 S.Ct. 2013 (1996), arguingagainst defendant'sright to introduceevidence of his voluntaryintoxication to establish diminishedresponsibility; Nunez v. Cityof San Diego,963 F.Supp. 912 (1995), supportingcurfew for under-eighteens;Johnson v. City ofDallas, 61 F.3d 442 (5thCir. 1995),supporting city laws against sleeping in public;Loperv. NYPD, 999 F.2d 699 (1993),supporting police action against panhandlers; and Michigan Department ofState Policev. Sitz,496 U.S. 444, 110 S.Ct. 2481 (1990), defendinghighway sobriety checkpoints. 3 Joycev. Cityof San Francisco,846 F. Supp. 843 (N.D. Calif.1994). 4 There is an excellentaccount of theSan FranciscoMatrix Program and thepublic responseto it in N. Wright,'Not in Anyone'sBackyard: Ending the "Contestof Nonresponsibility"and ImplementingLongterm Solutions to Homelessness'(1995) 2 GeoJ.on FightingPoverty 163 at 180-1. 5 J.Waldron, 'Homelessness and the Issue of Freedom' (1991) 39U.C.L.A. Law Rev. 295 [hereinafter'Homelessness']; reprinted in J. Waldron, Liberal Rights: Collected Papers 1981-1991(Cambridge: Cambridge University Press, 1993) [hereinafterLiberalRights]. 6 See especiallyRoulette v. Cityof Seattle, supra note 2, supportingcity laws against obstructionof sidewalk, and Johnsonv. City ofDallas, supra note 2, supporting city laws againstsleeping in public.

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions HOMELESSNESSAND COMMUNITY 373

Matrixprogram was an attempt,they said, to recoverthe public spaces of the city for the community.So long as homeless people remain en- camped in the city'sstreets, parks, and public squares, those places will be cluttered with tents, dirtysleeping bags, cardboard shelters, and stolen shopping carts and contaminated with urine, faeces, and drug paraphernalia. Such conditions,argued the AARR,make it verydifficult forordinary citizens, either individually or in families,to use those spaces in the way that theywere intended to be used. Panhandling, drinking, and variousforms of disturbedbehaviour exacerbate the problem, making the public urban environmentnot only unpleasant but hostile and potentiallydangerous. The result is that public places that used to be available to the whole communityare now 'becoming the preserve of those on the marginsof society.'7 The AARRargued thata communityhas a rightto controlbehaviour in its public spaces, and to outlaw activitiessuch as drinking,panhandling, sleeping on benches, washingin fountains,urinating and defecatingin public, and so on. The point of such restrictions,they said, is not to oppress the homeless or to diminishtheir liberty, but to reduce annoyance, to provide a fairbasis on which all citizens could make use of the public spaces of their city,and to allow parks and squares to become once again a healthy focus for the public life of the community.The AARRargued that communitiesbenefit from public spaces being kept sufficientlyattractive to act as public meetingplaces and as places where people voluntarilyspend their time.8The brief spoke eloquently of a time when citizensfrom all walksof lifecould spend theirleisure hours in public places, a time when parks and boulevards were places of 'interaction,integration, relaxation, and reflection.'9"And it urged the court to allow the cityto perseverein its effortsto restorethis communitarian mode of the use of its public spaces. It seemed to me importantto say,in response to these contentions, that although it is certainlytrue (as the AARRput it) that 'governments have the rightto regulate certain typesof conduct in public places, to ensure thatparks and sidewalksremain accessible and welcome to all,"' and although a citymust accept responsibilityfor maintaining the quality of its public places and may not treatwhat happens there as a matterof indifference,still the regulationof public space is a differentmatter in a communitysome of whose membershave no privatespace to retreatto than in a communityall of whose membershave access to privatespaces - homes- as well as public spaces, where theycan live theirlives and take

7 Amicusbrief ofAmerican Alliance for Rights and ResponsibilitiesinJoyce v. Cityof San Francisco,supra note 3 at 2 (on filewith author.) 8 See Teir,Restoring Order, supra note 1 at 256. 9 AARRbrief in Joyce, supra note 2 at 2. 10 Ibid.at 1.

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions 374 UNIVERSITY OF TORONTO LAWJOURNAL care of theirbodily needs. In the lattersociety, where everyonealso has a privateplace to go to, it is perhaps reasonable to say that the activities performedin public mightbe the complementof activitiesappropriately performedin private- a differentand complementaryset of activities. But in the formersociety, where some individualshave no choice but to live all theirlives in public, thatsame complementaritycannot prevail.I intend to elaborate that argument - about the implicit premise of complementaritybetween public and private- which I thinkunderlies much communitarianwriting (including, I fear, a certain amount of communitarianwriting by the left) about the use of public spaces, a little later in the essay.

III RobertEllickson

Before doing that, I would like to referto a second occasion that has stimulated my thoughts on these matters. This is a more academic occasion: the publication in 1996 in the YaleLawJournal of a long article by Yale law professorRobert Ellickson, entitled 'Controlling Chronic Misconduct in CitySpaces: Of Panhandlers,Skid Rows,and Public Space Zoning.'" ProfessorEllickson raised manyof the same issues as the AARR briefin Joyce;,but in a long law reviewpiece he was able to address the issues at greater length, to display the sinews of the argument,and to offerreaders a deeper opportunityfor reflection on the premisesof his account. (I should add that Ellickson is interested not only in the problem of regulatingpublic spaces but in the more abstractissues thatit raises,for example, issuesconcerning the emergence and effectivenessof social, as opposed to strictlylegal, norms.)'2 Let me stringtogether few quotations to giveyou a sense of the overall flavourof Ellickson's article:

In largecities in theUnited States, government owns as muchas 45% ofthe de- velopedland area and allocatemost of these public lands for use as streetsand highways....To socializeits members, any society, and especiallyone as diverseas the UnitedStates, requires venues where people of all backgroundscan rub elbows....A liberalsociety that aspires to ensure equality of opportunity and uni- versalpolitical participation must presumptively entitle every individual, even the humblest,to enter all transportationcorridors and open-accesspublic spaces.

11 (1996) 105 Yale L.J. 1165 [hereinafter'Controlling Chronic Misconduct']. 12 See also R.C. Ellickson, OrderWithout Law: How NeighborsSettle Disputes (Cambridge: Harvard UniversityPress, 1991) [hereinafterOrder Without Law].I discussthis aspect of Ellickson's work,in itsbearing on the issue ofstreet misconduct, infra in section XI.

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions HOMELESSNESSAND COMMUNITY 375

A space thatall can enter,however, is a space thateach is temptedto abuse. Societiestherefore impose rules-of-the-road forpublic spaces.... Rulesof proper streetbehavior are not an impedimentto freedom,but a foundationof it.... [T] o be trulypublic a space mustbe orderlyenough to invitethe entry of a large majorityof those who come to it.Just as disruptiveforces at a townmeeting may lower citizenattendance, chronic panhandlers, bench squattersand other disorderlypeople maydeter some citizens from gathering in theagora." The mediaare quickto report the gravest problems of the streets, such as armed robberies,drug trafficking, and driveby shootings. This Article [Ellickson says] focuseson problemsthat by comparison seem trivial:chronic street nuisances. Chronicstreet nuisances occur when a personregularly behaves in a publicpace in a waythat annoys - butno morethan annoys - mostother users, and persists in doingso overa protractedperiod."4 And Ellicksonoffers this definition of the phenomenon he takesas his targetin the article: A personperpetrates a chronicstreet nuisance by persistently acting in a public space in a mannerthat violates prevailing community standards of behaviorto the significantcumulative annoyance of persons of ordinary sensibility who use thesame spaces.15 Unless chronic street nuisances of this sort are controlled, says Ellickson,unless citydwellers can enjoy 'a basic minimumof decorum in downtownpublic spaces, theywill increasinglyflee fromthose locations to cyberspace,suburban malls, and private walled communities.''"6The rhetoricis more or less exactlythat of the AARR,cited by the court in Roulettev. Cityof Seattle.'As amicus American Alliance for Rights and Responsibilities explains on the city's behalf, "[a] downtown area becomes dangerous to pedestrian safetyand economic vitalitywhen individualsblock the public sidewalks,thereby causing a steadycycle of decline as residentsand touristsgo elsewhereto meet,shop and dine."'17 The reactions to Ellickson's article that I have seen's respond mostly to a particularpolicy proposal that he asks his readers to entertain: a

13 Ibid. at 1176. 14 Ibid. at 1168-9. 15 Ibid. at 1185 (emphasis omitted). 16 Ibid. at 1172. 17 Supra note 2 at 1430. 18 The best discussion to date is S.R. Munzer, 'Ellicksonon "ChronicMisconduct" in Urban Spaces:Of Panhandlers, Bench Squatters, and Day Laborers (1997) 32 Harv.C.R.-C.L.Law Rev. 1. I am gratefulto ProfessorMunzer for a most helpful conversation on these issues. See also M.D. Rosen, 'Our NonuniformConstitution: Geographical Variations of ConstitutionalRequirements in the Aid of Community' (1999) 77 Tex.L.Rev. 1129 at 1135n; A.J.Gold, 'The TrinityInitiative in Economic Perspective:Place or People

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions 376 UNIVERSITYOF TORONTO LAWJOURNAL systemof zoning in public places, permittingpanhandling and other offensivebehaviour at any timein certainzones thathe calls 'Red Zones,' permittingthem on an intermittentbut not chronic basis in other areas that he calls 'Yellow Zones,' and forbiddingthem altogether in those public places that he refersto as 'Green Zones.'9 I shall not say much about thatproposal, partlybecause it is not clear that Ellickson is offer- ing it as anythingother than a thought-experiment,20but mainlybecause I think it more important to ponder some of the premises than to evaluate the policy conclusions of Ellickson's account, for I have the impression that Ellickson's premises are shared by many who may not want to pursue his particularzoning proposal.

IV Costsand benefits

What Ellickson's article and the AARRargument have in common is concern about the offenceand annoyanceto ordinarycitizens caused by the presence and activitiesof homeless people in public places. (Profes- sor Ellickson himselfis not knownas a communitariantheorist; though there are clear communitarian themes in his work,21his ideological frameworkis more law-and-economics.)I want to begin mydiscussion by

Prosperity?'(1998) 30 Conn.L.Rev. 1317 at 1330n; R. Austin,'"NotJust for the Fun of It!": GovernmentalRestraints on Black Leisure, Social Inequality,and the Privatization of Public Space' (1998) 71 So.Calif.L.Rev. 667 at 705n; L. White, 'Searching for the Logic Behind WelfareReform' (1996) 6 U.C.L.A. Women's L.J.427 at439n; G.E. Frug, 'CityServices' (1998) 73 N.Y.U.L.R. 23 at 78n; and M. Foscarinis, 'Downward Spiral: Homelessness and its Criminalization' (1996) 14 Yale L.& Pol'y Rev. 1 at 3n. 19 Ellickson, 'Controlling Chronic Misconduct,' supra note 12 at 1220-6. The core insightof Ellickson's zoning proposal takes itsinspiration from the diversityof social norms traditionallyenforced in differentparts of American cities: Because demands on public spaces are highly diverse, city dwellers have historically tended to differentiatetheir rules of conduct for specific sidewalks,parks, and plazas. Some neighborhoods, like traditionalSkid Rows, have been set aside as safe harbors for disorderlypeople. Other sites,like tot-lots,have been allocated as refugesfor persons of delicate sensibility.A constitutional doctrine that compels a monolithic law of public spaces is as sillyas one that would compel a monolithic speed limit for all streets.(Ibid. at 1247.) 20 Ellickson introduces his zoning proposal by saying: 'As a mentalexperiment, imagine that it would be desirable for a city to have three codes, of varyingstringency, governingstreet behavior.' Ibid. at 1220 (myemphasis). This suggeststhat it is not to be takenseriously as an actual policyrecommendation. Elsewhere, however, Ellickson refersto the idea that 'a city'scodes of conduct should be allowed to varyspatially - from street to street,from park to park, from sidewalk to sidewalk' as a 'central normativethesis' of his article. Ibid. at 1171-2. 21 See ibid. at 1172, note 26. See also 'New InstitutionsFor Old Neighborhoods' (1998) 48 Duke L.J. 75 at 107 ff,defending neighbourhood-level communityorganization against liberal objections; and 'Propertyin Land' (1993) 102 Yale L.J. 1315 at 1344 ff, on communitarianarguments about property.

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions HOMELESSNESS AND COMMUNITY 377

consideringthe wayin whichhe categorizeswhat he calls 'annoyance' in an economic account. Early in the article,Ellickson attemptswhat appears to be an econo- mist's cost-benefitanalysis of chronic streetmisconduct. Both panhandling and bench-squattingare said to conferbenefits on various parties. As Ellicksonputs it,in the peculiar language used in analysesof thiskind, Panhandlersand at least some of theirdonors benefitfrom begging. The magnitudeof thesebenefits depends on the opportunitycosts incurred if the panhandlingwere to cease and bothpanhandlers and donors had to resortto theirnext best substitutes.22

For panhandlers, the benefitis the differencebetween begging and (say) collectingcans and bottles; forthose donors who 'take affirmative pleasure in satisfyinga request fora handout,' the extent of the benefit turns on 'the quality of the almsgivers'other alternativesfor being charitable.'23Similarly, bench-squatting has economic benefits:

[T]he magnitudedepends on the qualityof the bench squatters'snext best alternatives.These mightinclude: squattingin anotherpublic locale better suited to long-termstays; cycling among a number of public places (...); spend- ing moredaylight hours indoors (perhaps in a board-and-carefacility, a drop-in center,a rentedapartment, or a relative'shome); and voluntarilyinitiating insti- tutionalization.Because the first alternatives listed are closesubstitutes, the benefitsof an entitlementto bench squatin a particularlocation ... are apt to be small.24

Those are the benefits.What about the harms?The harms,Ellickson says,consist in 'minor annoyance' multipliedover many individualsand sustained over a period of time. 'When being pan-handled,' he says, 'a pedestrianof ordinarysensibility may feel some combinationof: aggrava- tion [sic] that his privacyhas been disturbed,resentment that the panhandler's plea has a high probabilityof being fraudulent,and fear.'25 The fear element may not be present among those pedestrians who encounter chronic panhandlers who are familiarto them; however, [i]n otherrespects, ... theencounter may be moreannoying than an encounter withan unfamiliarpanhandler, A pedestrianwho sees a regularpanhandler is likelyto become increasinglyirked that the supplicant has notsought aid from charitiesand welfareagencies better able thanpedestrians to appraisedesert.26

22 'Controlling Chronic Misconduct,' supra note 12 at 1179. 23 Ibid. at 1179-80. 24 Ibid. at 1183. (I make no comment on the plausibility- or tone - of this analysis.) 25 Ibid. at 1181. 26 Ibid. at 1182.

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions 378 UNIVERSITYOF TORONTO IAWJOURNAL

And such a panhandler may also annoy because his activitysignals to the pedestrianthe breakdownof a social norm- such as the workethic - that the pedestrian cherishes.27As for the annoyance caused by bench- squatting,Ellickson saysthe following: The mostflagrant examples involve offense to a numberof the senses.A man who sitsin a well-traffickedspace amid shoppingcarts full of junk, who stinks withbody odor, and whourinates publicly into plastic jugs, is likely to triggerfre- quent complaintsto the police.A womanwho sleepson a busysidewalk, who smells of feces, and who shoutsobscenities certainly engages in offensive behavior....28

Now, these may all seem trivialharms - annoyances rather than injuries. But Ellickson is rightto emphasize that a set of harms should not be ignored in policyanalysis merely because each one considered by itselfis verysmall. As the moral philosopher Derek Parfithas emphasized, and as those who study collective action problems know, consequentialistanalysis can go seriouslywrong by ignoring tiny harms.29 An individualautomobile drivermay release onlysmall amounts of toxic wastes into the environmentwhen he drivesto work.But the effectsof hundreds of thousands of driversusing thatfreeway on a given day may add up to deadly pollution. Moreover,it maybe impossibleto solve such a pollution problem except by subjectingall motoriststo certain regulations (e.g., requiring smog checks) and punishing them one by one - despite the almost imperceptibleharm thateach causes individually- if theydo not complywith the regulations.So it is perfectlyfair for Ellick- son's analysisto take account of the fact that because a piece of street misconduct (relativelytrivial in itself) 'occurs in a public place, it may affecthundreds or thousands of people per hour,' and forhim to insist, too, that 'as hours blend into days and weeks, the total annoyance accumulates.'3oNo doubt there are furtherquestions to be asked about how much moral weightwe should give to the aggregationof tinyharms in a social calculus. Do we everwant to say,for example, thata verygreat harm (say, torture) inflictedon a single individual is the equivalent of the sum of tinyharms (analogous to those Ellicksonis studying)suffered byN individuals(where N is a verylarge number,say in the millions)?31 If the answer is 'No,' then can a somewhat less serious individual harm, such as arrest,suffered by one person be outweighedby (say) thousands of ambient annoyancessuffered by Ellickson's passing pedestrians? If the

27 Ibid. 28 Ibid. at 1183. 29 See D. Parfit,Reasons and Persons ch. 3 (Oxford:Clarendon Press, 1984). 30 ControllingChronic Misconduct, supra note 12 at 1177. 31 See Parfit,supra note 31 at 75-82.

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions HOMELESSNESS AND COMMUNITY 379 answerto the latterquestion is supposed to be 'Yes,' then what's the principle underlyingthe difference?These are all interestingquestions, and theygo to the heart of some central difficultiesin utilitariananalysis.32 But the problem in Ellickson's analysisthat I would like to explore arises at an earlier stage.

v Aredistress and annoyanceharms?

The question I want to ask is whetherthe discomfortcaused to ordinary pedestrians by the presence and activitiesof homeless people should even be considered a harm at all. On the face of it,the question seems preposterous.Of course discom- fortand annoyance are harms. They may not be veryserious harms - except to the extent that they accumulate - but surelythey are to be counted on the debit side of any plausible utilitariancalculus. I am not so sure. Think of it along these lines. Imagine that in a countrywhere homelessness has only recentlybecome a problem, a citizen previouslyunaware of the extent of the povertyin his society comes across a person livingon the streetsin filthand squalor. He is likelyto be distressedby the spectacle, but his distressmay have the followingflavour: 'This is awful.I am glad I have found out about this,' and he maybe moved byhis horrorto tryand do somethingabout it.33Is thisdistress a harmto the citizen,something that in a utilitariancalculus should count protanto against his findingout about povertyand in favour of his being sheltered fromhis knowledge?I thinkwe should say not- not even thatit is a slightharm out-balancedperhaps by greaterbenefits associated withhis knowledge. On the contrary,there is a clear sense in which distressoccasioned by the spectacle of another's sufferingis a good rather than an evil (just as pleasure occasioned by another's sufferingis an evil,not a good). If there is first-ordersuffering of thissort around, then it is betterthat it be seen and thatpeople be distressedby it than that it remain invisibleto all but the immediate sufferers.A world that differedfrom our own only in that the spectatorsof such suffering were not moved to any sortof distressby the spectacle of povertywould

32 See also the discussion of trade-offsinJ. Waldron, 'Rightsin Conflict'in LiberalRights, supra note 5 at 208-11. 33 Cf. W. Shakespeare, King Lear,Act III, scene 4: Lear. Poor naked wretches,whereso'er you are, That bide the pelting of thispitiless storm, How shall your houseless heads and unfed sides, Your loop'd and window'd raggedness,defend you From seasons such as these? 0, I have ta'en Too littlecare of this!

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions 380 UNIVERSITYOF TORONTO IAWJOURNAL be pro tantoa worse world. I shall call this the 'Appropriate Distress Argument.' Or thinkof the situationalong these lines. A pedestriancomes across a chronic bench-squatter,and he feels distressof the followingkind: he thinksto himself,furiously and agitatedly,'It is outrageousthat people like thisshould sit idlyaround, instead of fulfillingtheir moral obligation to get ajob and contributeto the social product.' But ifsomeone asked why the bench-squatterhas a moral obligation to get ajob, itwould surelybe inappropriateto cite the pedestrian's agitationas a utilitarianreason - that is, 'People ought to get jobs so as to avoid distressingthose who think that they ought to get jobs.' To argue in that way would be like arguingin a circle; one would be tryingto defend a moral propositionby citing formsof distresswhose occurrence presupposed that the moral propositionwas true.I shall call thisthe 'ExternalPreference Argument.' Both arguments seem applicable to Ellickson's analysis - I mean applicable as critiques.In both regards,his ratheringenuous account of the 'cost' or the 'harm' caused by chronic misconductin public places ignores a lot of quite interestingwork that has been done in recent moral philosophyabout theway in whichpreferences and putativeharms ought to be counted in a social calculus.

VI TheAppropriate Distress Argument

There is a well-knowncontroversy about the application of the 'Harm Principle' propounded inJohn StuartMill's book On Libertythat goes as follows. Suppose someone is distressedby the conduct of another - suppose forexample thata person feels offenceand revulsionwhenever he sees twomen kissing.Is distressof thiskind - whichis certainlypainful to experience - to be counted as harmfor the purposes of a principle which holds that 'the only purpose for which power can rightfullybe exercised over any member of a civilizedcommunity, against his will,is to preventharm to others'?34 The most plausible answer to thatquestion - which I defended some time ago in a littlepiece called 'Mill and the Value of Moral Distress'35 - is emphatically negative. For the purposes of Mill's argument in On Liberty,distress of this kind is to be counted as a social good.This is not because it is intrinsicallygood forhomophobes to suffer,but because it is importantfor people to be confrontedwith ideas and waysof life that challenge theirown comfortablepreconceptions. Ethical confrontation - the confrontationof ideas and lifestyles- is a positivegood forMill. But of course it is not a painless business.It hurtsto be confrontedin debate

34 J.S. Mill, On Liberty,ed. C.V. Shields (Indianapolis: Bobbs Merrill,1956) at 13. 35 (1987) 35 Pol.Studies 410, reprintedin LiberalRights, supra note 5 at 115.

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions HOMELESSNESSAND COMMUNITY 381 ifone takesseriously the viewsone is propounding,and it is distressingto be faced withexamples of lifestylesthat pose a challenge to the foundations of one's own. People are naturallyupset when theyare involvedin this sort of confrontation.If no one is disturbed,hurt, or distressedin thisway, that is a sign that vigorousethical confrontationis not taking place, and, for Mill, that in turn is a sign that the intellectual life and ethical and culturalprogress of our civilizationmay be grindingto a halt. This is whyMill is verysuspicious of any proviso to the effect'that the freeexpression of all opinions should be permittedon conditionthat the manner be temperate':

If thetest be offenceto thosewhose opinions are attacked,I thinkexperience testifiesthat this offence is givenwhenever the attack is tellingand powerful,and thatevery opponent who pushes them hard, and whomthey find it difficultto answer,appears to them,if he shows any strongfeeling on the subject,an intemperateopponent.3"

Intemperance and the offence it inevitablyoccasions are indispens- able forconfrontation and forthe progressthat only confrontationcan bringabout. Think whatwould be entailed byan interpretationthat did regard thissort of offenceas sufficientto cross the thresholdrequired by the Harm Principle. On thatassumption, what ought to be taken as evidence that freedom of thought and lifestylewas promoting progress would be invokedinstead as a primafacie reason forinterfering with that freedom.A sign of vitalitywould be cited as a necessarycondition forthe legitimate suppression of that vitality.A symptomof progress would count as ajustificationfor acting in a waythat would bringprogress to a halt. Mill cannot have held such a view. Against this, it will be said that Mill was a utilitarian.How can a utilitariannot count discomfort,revulsion, and distressas pain?37How can a utilitariannot count them - as ProfessorEllickson counts them - on the debit or cost side of the social calculus? To answer this,we must bear in mind the sortof utilitarianMill was. Mill's utilitarianismis not a Benthamite calculus of pleasures and pains, or of satisfactionsand dissatisfactions,of all sorts.The value on which liberty(defended by the Harm Principle) is based is certainlyutility, on Mill's account; but, as he insistsin the Introductionto On Liberty,'it mustbe utilityin the largest sense, grounded on the permanent interestsof man as a progressive being.'38And I take it thatthis passage refersnot merelyto the nature of

36 Mill, supra note 36 at 64-5. 37 This objection is put forwardby T. Honderich, 'On Libertyand Morality-Dependent Harms' (1982) 30 Pol.Studies 507. 38 Mill, supra note 36 at 15.

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions 382 UNIVERSITY OF TORONTO LAWJOURNAL

Mill's utilitariancomputations - for example, taking a long-runrather than a short-runview - but also to the content and character of his utilitarianvalues. If we accept the argumentsabout the importance of progressin Chapter 2 of On Liberty,or the argumentsabout spontaneity in Chapter 3, it is not open to us to say that the distressexperienced when one's preconceptionsare challenged goes againstone's interestsas a progressivebeing. A creaturewho defined his interests- even in part- in terms of being free from the shock of ethical debate or free from anxietyabout the grounds and worthof his lifestylewould be like the satisfied'fool' in Mill's Utilitarianism.39 Now, I am not sayingthat Ellickson's pedestriansactually benefit by being confrontedwith the unusual lifestylesof bench squattersand so on (though they probably do). I have invoked Mill's argument primarily because it opens up forconsideration an arrayof waysin whichwe might reject Ellickson's rathersimple-minded characterization of all his pedestrians'distresses as harms. The particularargument I want to make is thatif Ellickson's pedestrians are regarded as progressivebeings, in Mill's sense, then we should not go around sayingthat they have an interestin not knowingor in not perceiving the true state of affairsin their society- the condition in which some of their fellowcitizens are having to live - simplybecause that knowledge or perception is distressingto them. If the situationof some in society is distressing,then it is important that others be distressedby it; if the situationof some in societyis discomforting,then it is importantthat others be discomforted.40 (By the same token, the distressoccasioned by finding out some unpleasant fact of our historyshould not be counted as a harm in a sophisticatedsocial calculus; it seems crazyto say,as the tenor of Ellick- son's analysiswould seem to imply,that people of decent sensibilityare harmed by their distressingknowledge of the historical existence of slavery,for example, or genocide.)41 To put it another way,if the basis of our moral sense is, as Rousseau put it, a natural repugnance at seeing a fellow being suffer,42then it is

39 'Utilitarianism' in J.S. Mill, Dissertationsand Discussions:Political, Philosophical and Historical,vol. 3 (New York: Henry Holt & Co., 1882) 300 at 312. 40 Thus I agree withMunzer, supra note 20 at 33, when he writes:'Seeing the distressof those who panhandle or bench squat is not, or at least not merely,an annoyance. It is rather,or also, an experience of gettinginformation, which no rational person should consider a social harm,about the lives of a significantfraction of the poor in America.' 41 But compare R.K. Fullinwider, 'Patriotic History' in R.K. Fullinwider,ed., Public Education in a MulticulturalSociety: Policy, Theory Critique (Cambridge: Cambridge UniversityPress, 1996) 203. 42 J.-J.Rousseau, 'Preface' inJ.-J.Rousseau, A Discourseon theOrigin ofInequality, trans. P. Coleman (New York: Oxford UniversityPress, 1994) 3.

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions HOMELESSNESS AND COMMUNITY 383 importantthat such repugnance not be treated as itselfan evil on the order of the first-ordersuffering that evokes it. Or, at the veryleast, it is importantthat it not be treatedas an independentevil, that is, as an evil that is independentlyrelievable or preventable. The only appropriate way to relieve or prevent the sufferingoccasioned in me by seeing another being suffer(which we may call 'respondent suffering')is to relieve or preventthe first-ordersuffering to which it is a response. If,on the contrary,one were to tryto relieve or preventrespondent suffering as an independent matter,one mightdo one's best to ensure that first- order sufferingtook place as faras possible out of sight,unperceived by those who mightbe moved by it. (Or one would try,by education or conditioning,to deaden the affectiveresponse in people to such first- order distress as remained visible.) But an attempt so motivated to render first-ordersuffering invisible, or to block the normal response to it,would be a wayof showingthat one did not understand the function of respondentsuffering and empatheticdistress in a human being. And my point is precisely that Ellickson's categorization of any negative feeling as harm,in the encounters he considers, shows preciselysuch a misunderstanding- a misunderstandingthat is not, I think,mitigated by the use of such jargon termsas 'compassion fatigue.'43 Having said that,in all fairness,I should add thatsympathetic distress at the sightof the homeless is not Ellickson'smain concern in the article so far as the putativecosts of streetmisconduct are concerned. He also mentions revulsionat body odours and the stinkof urine and faeces - and annoyance. So far as revulsionat odours is concerned, the situation is compli- cated. It is, of course, appropriate to recoil in public fromthe smell of urine and faeces. This sort of sensorydistress has a proper motivating effect,namely to ensure as faras possible thathuman waste is disposed of in sanitaryconditions. The argumentfor providingpublic lavatoriesis partlydependent on the importance of avoiding distressof this kind.44 But so far as the homeless are concerned, that is only part of the argu-

43 E.g., ControllingChronic Misconduct, supra note 12 at 1168, 1178, 1218; the same language is also used bythe AARR, e.g., in Teir, 'MaintainingSafety and Civility,'supra note 1 at 288. For a critique of the use of thisphrase, see N. A. Millich, 'Compassion Fatigue and the First Amendment: Are the Homeless Constitutional Castaways?' (1994) 27 U.C.DAVIS L.R. 255. 44 For discussion of the lack of provisionof public lavatoryfacilities in the United States and itsimpact on the homeless, see M. Davis, 'A Logic like Hell's: Being Homeless in Los Angeles' (1991) 39 U.C.L.A.Law Rev. 325 at 329-30; M. Foscarinis, K. Cunningham-Bowers,& KIE. Brown, 'Out of Sight- Out of Mind? The Continuing Trend Toward The Criminalization of Homelessness' (1999) 6 GeorgetownJ. on Poverty Law & Policy 145 at 154; and L. Sossin, 'The Criminalization and Administrationof the Homeless: Notes on the Possibilitiesand Limitsof Bureaucratic Engagement' (1996) 22 N.Y.U.Rev.L.& Soc.Change 623 at 653.

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions 384 UNIVERSITY OF TORONTO LAWJOURNAL ment. As I emphasized in myearlier piece, the main argumentfor this provisionhas to do withthe immediate distressand indignitysuffered by those who have to live almost theirwhole livesin a public space in which such facilitiesare not provided.45I will returnto thislater.46 For now, the point I want to emphasize is thatsince a conscious political decision has been made notto providesuch facilities- thisis true in the United States (for reasons, I mustsay, that entirely escape me);47 it is not true in most other advanced societies- then it is surelya bit much forthose who have failed to offereven thismodicum of social provisionto the homeless to complain about the consequent smell and to treatthat as grounds for offeringthe homelessfurther indignities and restrictions.

VII Cognitivedissonance

What about annoyance? I thinkmuch of the angryside of the discomfi- ture that concerns Ellickson is a matterof cognitivedissonance, associated withthe visiblerefutation of the claims thatAmericans are proud to make about theirsociety.48 They say that theirsis a just and prosperous society,a societyof equal opportunity;they say that capitalismand the market economy work wonderfully.But they are confronted on the streetsof theircities with hundreds or thousandsof persons who live on the verymargins of civilizedexistence, on the verymargins of lifeitself - who flauntin theirpersons the abject and desperate povertythat disfig- ures American societyand the idea of the American dream. Since their feeling good about themselvesas Americans depends primarilyon a sense that thesethings cannot be, ordinary people tend to respond with annoyance and anger to the sightthat they very evidently are. Now, I am not making the argument that the United States should live up to its ideals - that is almost certainlya hopeless enterprise.But political and social theories that count as harm - for purposes of social policy - the negativefeelings of anger and annoyance thatpeople experience when theyare shown that theirsociety is not livingup to its ideals - theories that count this as harm, to be avoided if possible by removing the spectacle thatoccasions the dissonance - are treadingon verydangerous ground. Since the Enlightenment,it has been a principle of good social

45 Waldron, 'Homelessness,' supra note 5 at 320-1 (LiberalRights at 334-5). 46 See infrasection X, especially notes 89-91 and accompanying text. 47 See C. Habermas, 'CityHall Can't AnswerNature's Call' New YorkTimes (11 February 2000) Bi; A. Hagedorn & W. Green, 'AdvocacyGroup-Sues NewYork, Citing Lack of Public Restrooms' WallStreet Journal (1 November 1990) B8. See also Lucasv. Dinkins, 608 N.Y.S.2d 403 (App.Div. 1994), holding thathomeless individuals'complaint about New York City'sfailure to provide public toiletswas notjusticiable. 48 For a similar phenomenon in Canada, see J. Stackhouse, 'My Life Withouta Home' The [Toronto] Globe and Mail (18 December 1999) 'Focus' 1, and the ensuing discussion.

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions HOMELESSNESS AND COMMUNITY 385 and political theorythat the social order should be subjectto a constraint of transparency- which means, roughly,that social stabilityshould not depend on any comprehensivemisunderstanding of social realityon the part of citizens.49In a well-orderedsociety, asJohn Rawlsputs it,'nothing need be hidden.'"5Now, of course,we all knowthat, in the United States at least,social stabilitydoes largely depend on misunderstandingand false ideological perceptionof actuallyexisting social conditions.But it is rare to find a theoristmaking a virtueof it, let alone designatingthe annoyance at havingthe illusiondispelled as a cost or a harm,to be prevented, ifneed be, by the police and the criminallaw. Of course, Ellickson would not describe it in this way. He resorts instead to the simple utilitarianformulas of the law-and-economics movement.The flawin his articleis thathe has not thoughtthrough the significanceof lumping all formsof distresstogether as social harms. Law-and-economicstypes often present themselvesas hard-headed men of the world,willing to take a much clearer viewof whatis reallygoing on than most of theirsentimental philosophical opponents.5' But thisself- presentationis undermined ifwhat law-and-economicsyields is a social theorythat panders, in the name of utilityor of the minimizationof social cost,to people's desiresnot to have theirillusions about the society theylive in dispelled.

VIII Brokenwindows

This may be the appropriate place to say somethingabout the relation between the visibilityof homelessness and a theory of community policing sometimesreferred to as the 'Broken WindowsTheory.'52 This is

49 SeeJ. Rawls,A TheoryofJustice (Cambridge: Harvard UniversityPress, 1971) 133 at 454. See also the discussioninJ. Waldron, 'Theoretical Foundations of Liberalism' (1987) 37 Phil.Q. 127 at 134-5 and 146 ff.,reprinted in LiberalRights, supra note 5, 35 at 43-4 and 57 ff. 50 J. Rawls,Political Liberalism (NewYork: Columbia UniversityPress, 1983) at 68. Rawls adds in a note: ' ... in a freesociety that all correctlyrecognize as just there is no need forthe illusionsand delusions of ideologyfor society to workproperly and forcitizens to accept it willingly.In thissense a well-orderedsociety may lack ideological or false consciousness.' Ibid. at 68-9, note 21. 51 It was, afterall, the grandfatherof rational choice theory,Thomas Hobbes, who invented and defended the principle of transparency.See Hobbes, Leviathan, ed. RichardTuck (Cambridge: CambridgeUniversity Press, 1991) ch. 30 at 231-2. See also J. Waldron, 'Hobbes on Truth and Civil Doctrine' in A. Rorty,ed., Philosopherson Education(London: Routledge, 1998), and 'Hobbes and the Principle of Publicity'in B. Honig, ed., Essaysin Honorof Richard Flathman (forthcoming, 2001). 52 For good overviewssee D. Livingston,'Police Discretion and the Quality of Life in Public Places: Courts,Communities and New Policing' (1997) 97 Colum.L.Rev. 551; B.E. Harcourt, 'Reflecting on the Subject: A Critique of the Social Influence

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions 386 UNIVERSITY OF TORONTO LAWJOURNAL the viewput forwardby George KellingandJames Q. Wilson in a famous article published in 1982, in which theyargued thatvisible signs on the streetsof lack of repair or lack of community concern - signs like a broken window,a pile of uncollected garbage, abandoned furniture,or an abandoned car - are likelyto attractfurther disorder, and also crime, because theysignify to anyone who is interestedthat social controlsare weak or attenuatedat thatlocale."5 According to some proponentsof this theory,derelict human beings behaving in an offensivefashion are just one more such sign of disorder:

[J]ustas unrepairedbroken windows in buildingsmay signal that nobody cares and lead to additionalvandalism and damage,so untendeddisorderly behavior may also communicatethat nobody cares (or thatnobody can or will do anythingabout disorder) and thuslead to increasinglyaggressive criminal and dangerouspredatory behavior."54

On thisaccount, 'the ill-smellingdrunk [or] the unchecked panhandler is, in effect,the firstbroken window.'55Or, as Ellickson puts it, 'a regular beggar is like an unrepaired broken window - a sign of the absence of effectivesocial-control mechanisms in thatpublic space.'56 Now, I thinkwe ought to be verywatchful of theoriesof community control that assimilatehuman beings even figurativelyto things,like uncollected garbage or shatteredwindow panes. (If Kellingand Wilson had titledtheir article 'Broken People,' would itperhaps have had a different impact?) If we are to countenance policy talk about broken people on the analogyof brokenwindows, then we need to ask twotough questions: (1) Relative to what norms of order are bench-squattersor panhandlers or smellystreet people described as 'signs of disorder'?Are these the norms of order for a complacent and self-righteoussociety, whose more prosperous members are tryingdesperately to sustain various delusions about the situationof the poor? Or are theynorms of order for a societywhose members are attemptingin good faithto live honestly with a given mixture of great prosperityand great poverty?In some societies whose membersare familiarwith great and endemic poverty- India springsto mind as an example - people regard streetbegging as a normalactivity, and not at all as a disorder.Their response to any com-

Conception of Deterrence, the Broken Windows Theory, and Order-Maintenance Policing NewYork Style' (1998) 97 Mich.L.Rev. 291. 53 J.Q. Wilson & G.L. Kelling, 'Broken Windows' AtlanticMonthly (March 1982) 29. 54 G.L. Kelling, 'Acquiring a Taste for Order: The Communityand Police' (1987) 33 Crim.& Delin. 90 at 92, quoted byAARR counsel Rob Teir in 'MaintainingSafety and Civility,'supra note 1 at 290. 55 Kelling & Wilson, supra note 55 at 34. 56 'Controlling Chronic Misconduct,' supra note 12 at 1182.

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions HOMELESSNESS AND COMMUNITY 387 plaint about such 'disorder' would be: 'Given the existence of poverty (whichyou propose to do nothingabout), whaton earth do you expect?' I belieye there is a dignityin that candour, certainlycompared to the shamefuland self-righteousdenial thatwe findon the streetsof America and the infantile preference for image over reality that dominates discussion in the United States. So the firstquestion is this: Are the norms - relative to which human windowsseem 'broken' - norms of order forsociety in which it is envisaged thateveryone has a home to go to, thatis, normspredicated on the assumptionof complementaritythat I mentioned in section III? Or are they norms of order based on an honest grasp of economic realityin an unequal society? (2) The second question is this:Relative to whatevernorms are appropriate,what (according to the Kelling and Wilson approach) is to count as fixingthe window,when the 'broken window' is a human being? A panhandler needs, or thinkshe needs, money. Presumably,giving him money is not 'fixing the window' in terms of the Broken Windows Theory.But then whatis? Suppose we are regularlyconfronted in a dere- lictneighbourhood withpeople who smell of urine and faeces. There's a brokenwindow, according to the Kellingand Wilson approach. Now how do we fix it? For some reason - in the United States at least - nobody (certainlynobody who invokesthis theory) seems to thinkthat the appropriate answer is: 'Provide public lavatoriesand public shower facilities.' Instead, fixingthe windowis taken to mean roustingthe smellyindivid- ual and making him move out of the public park or citysquare. (It is as though the smartestway to fixan actual broken windowwere to knock down the whole building,or move it tojust outside the edge of town.) There is much more one could say about the Broken Windows Theory. One could question, for example, the theory's deterministic assumptionof there being a constantpotential for escalating, predatory crime, roaming the cities,looking for disorderlyand neglected sites to settle on. One could talk about the way in which Ellickson, Teir, and others have converted (or, more accurately,hijacked) a theoryabout policing priorities57and turned it into a theoryof legislation- that is, into a theoryabout what sorts of things(e.g., panhandling, sleeping in public) should be made into offences. And, in that regard, one could dwell for a moment or two on the ethics of criminalizingactivities performedby one group of people in order to offsetthe attractionthat theirderelict condition mayoffer to other offenders."8

57 That is, a theoryabout the importance of the police focusing on minor as well as major crime: see, e.g.,W.J. Bratton, 'The New York CityPolice Department's Civil Enforcementof Quality-of-LifeCrimes' (1995) 42J. Law & Pol'y 447. 58 Cf. Foscarinis et al, supra note 46 at 153: 'This punitive approach raises serious concerns about fundamentalfairness.... [P]unishing one group of people to prevent

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions 388 UNIVERSITY OF TORONTO LAWJOURNAL

Mainly, however, what I want to say about the Broken Windows Theory is the following.Prosperous societies in the West,particularly the United States (but now, increasingly,Canada, the European Union, Australia,and New Zealand as well) have entered into a bargain withthe devil. For decades we conjectured that povertyfor some would lead eventuallyto a deteriorationin the qualityof lifefor everyone, even for the rich and comfortable.And on the basis of thatconjecture we sought to mitigatethe worsteffects of inequality.We did so in our own interest, as well as on the basis of more altruisticand social justice concerns. We believed that if we didn't, inequalitywould eventuallyredound to the detrimentof us all. Since 1980, however,the United Kingdomfirst, then the United States, and then other countriesfollowing their lead have decided to test that conjecture and, if possible, refute it. (This is the devil's bargain I referto.) Now we are workingon a differenthypothesis: maybe extreme povertyfor a substantialsection of societycan be toler- ated withimpunity, without undermining (even in the long run) security and qualityof life for the most prosperous and the opportunitiesthey cherishfor their children. Maybe aggressivepolicing strategies mean that we can have all the glamourof a prosperous-lookingsocietywithout doing verymuch - doing perhaps much less than we have done in the past - to help the poor, the unfortunate,and those who have made disastrous choices. And who knows? Maybe that hypothesisis true. Maybe the bargain with the devil will work, with a huge payoff - a sort of compassion-fatiguedividend - forthose of us who are prospering. Maybe. But it does seem a bit much - it seems unconscionable, in fact - to complain about and stigmatizethe poorest of the poor forrefusing to cooperate with this experiment in the cosmetics of injustice, for refusingto play theirpart in thisnew scenario in which we tolerate and benefitfrom inequality while protectingour illusionsand sensibilities.It does seem a bit much to characterizethe poorest of the poor as 'broken windows' relativeto a self-imageof righteousprosperity and order,when it is not entirelyclear - indeed, when it is exactly the point of our experimentto see - whetherwe are entitledto project thatimage."5

futurecriminal activity by others runs afoul of the basic notions of equity underlying our criminaljustice system.' 59 In times past, inequalityand great povertyhave themselvesbeen regarded as social disorders,as 'broken windows' in the fabricof a well-orderedsociety; and it has been thought that sweeping them from view,far fromremedying the disorder, actually compounds it. Here I agree withMunzer, supra note 21 at 33, when he writes,'The existence of whatEllickson calls "streetdisorder" has not merelya corrosiveeffect on the public good of harmonious citylife but also is an indicator of social injustice. It serves as a constant reminder thatsomething is deeply wrongwith a societythat has vast numbers of bench squattersand chronic panhandlers.'

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions HOMELESSNESS AND COMMUNITY 389

IX TheExternal Preference Argument

I mentioned twolines of argumentagainst Ellickson's characterization of offence and annoyance as social costs. The firstwas the 'Appropriate Distress' argument,which I set out in section VII. The second line of argumentI called the 'ExternalPreference Argument,' and it pertainsto Ellickson's claim that '[a] pedestrianwho sees a regular panhandler is likelyto become increasinglyirked that the supplicanthas not soughtaid from charities and welfare agencies better able than pedestrians to appraise desert'" and to his view that this 'irk'or thisbeing irkedconsti- tutesa harm to the 'irkee.' The External Preference Argumentwas made originallyby Ronald Dworkin to criticize versions of the utilitariancalculus that treated people's preferencesas to what othersshould do or have (which Dworkin called 'external preferences') on a par with people's preferences for themselves ('personal preferences')."6In the debate about segregated education, for example, Dworkin believed that it was appropriate to consider both the costsand the benefitsof integration,but he thoughtit was not appropriate to count, as a cost of integration,the frustrationof the desire of some raciststhat blacks count forless or thatthey be denied equal educational opportunities. Counting such preferences would undermine whatevermoral appeal utilitariancalculations have as a fair basis forthe evaluation of costs and benefits.One ground for the moral appeal of the utilitariancalculus is the utilitarian's commitment to equality,at least at the level of inputs:each person is to count for one, nobody formore than one. If a utilitarianargument counts external preferences along withpersonal preferences, then the egalitarian character of that argument is cor- rupted, because the chance that anyone's preferenceshave to succeed will then depend, not on the demands that the personal preferencesof others make on scarce resources,but on the respect or affectionthey have for him or his way of life. If external preferencestip the balance, then the factthat a policymakes the communitybetter off in a utilitarian sense would notprovide ajustificationcompatible with the rightof those it disadvantagesto be treatedas equals."62 I will not go into what Dworkinhas made of thispoint or the central role it plays in his theory of rights-as-trumps-over-utility."3I do want to

60 ControllingChronic Misconduct, supra note 13 at 1182. 61 R. Dworkin, TakingRights Seriously, rev. ed. (London: Duckworth,1977) at 234 ff.and 275-6. 62 Ibid. at 235. 63 But see ibid. at 274-8 and 356-9. The connection is set out inJ. Waldron, 'Pildes on Dworkin's Theory of Rights' (2000) 29J.Legal Stud. 301.

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions 390 UNIVERSITY OF TORONTO LAWJOURNAL emphasize, however,that it applies not only to vicious external preferences like the racistonesjust mentioned but to virtuous,even admirable, external preferencesas well. In a notable critique of Dworkin's argument, H.L.A. Hart argued that it was in fact perfectlyappropriate for external preferences to be counted in the determination of social policy."4Hart argued that the importance of not disqualifyingexternal preferences is shown by the example of homosexual law reform in England in the 1960s,where, he says,it is perfectlypossible that'it was the disinterestedexternal preferencesof liberal heterosexuals that homo- sexuals should have this freedom that tipped the balance against the external preferencesof other heterosexualswho would deny this freedom.'"6 How could anyonepossibly complain about countingexternal preferences in that case? And ifwe count them in thatcase, whynot count them in anycase, includingcases of the sortthat Ellickson describes? Hart's critique has convinced many commentators.66But I think Dworkin'sresponse to Hart is exactlyright, and it is adaptable to the case we are considering. Dworkin concedes that a simple-mindedform of utilitarianismwould count external preferences: it 'would count the attractivepolitical convictions of the liberals of the nineteen-sixties simplyas data, to be balanced against the less attractiveconvictions of others, to see which carried the day in the contest of number and intensity.'"67But this,he says, is surelynot how the liberals of the 1960s intended theirpreferences to be taken:

They of course expressedtheir own politicalpreferences in theirvotes and arguments,but theydid not appealto the popularityof thesepreferences as providingan argumentin itselffor what theywanted, as the unrestricted utilitarianargument I opposewould have encouraged them to do."8

The arguments made by the liberal supportersof gay rights were presumablyarguments of principle- arguments(e.g., about libertyand privacy)intended to stand on theirown grounds,and not on the mere

64 H.L.A. Hart, 'Between Utilityand Rights' in A. Ryan,ed., TheIdea ofFreedom:Essays in Honor ofIsaiah Berlin(Oxford: Oxford UniversityPress, 1979) 77 at 86 ff. 65 Ibid. at 92-3. 66 See, e.g.,J.H. Ely, 'Professor Dworkin's External/personal Preference Distinction' (1983) Duke L.J. 959; T. Morawetz, 'Persons Without History:Liberal Theory and Human Experience' (1986) 66 B.U.L.Rev. 1013 at 1017; and P.E. Johnson, 'Do You SincerelyWant to Be Radical?' (1984) 36 StanfordL.Rev. 247 at 276. However, for a more careful view of Hart's critique, see J. Raz, 'Liberalism, Skepticism, and Democracy' (1989) 74 Iowa L.R. 761 at 772-4. 67 R. Dworkin,'Rights as Trumps' inJ. Waldron, ed., Theoriesof Rights (Oxford: Oxford UniversityPress, 1984) 153 at 162. 68 Ibid.

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions HOMELESSNESS AND COMMUNITY 391

factof theirbeing held and supported.Certainly Dworkin is not arguing thatpeople should refrainfrom expressing their external preferences or theirprincipled convictions;he is not urginga politics of self-interest."6 His argumentis, rather,that the utilitariancalculus has specificwork to do in politics, and that when it does its work it should work in the domain of personal preferencesto establishwho would be harmed or who would be benefitedin termsof what theywant for themselvesfrom some politicalproposal. That, then,may be used as a wayof defendingor criticizingpolitical proposals, and it may be appealed to as a basis for voting one way or the other.7"But it can do thiswork only if it is con- ceived as a distinctform of politicaljustification, not if it operates as a functionover moral convictionsthat people already have and that they justify(if theydo) on independent grounds. The point, then, is this.In cases where people feel vehementlyabout some matterof principle- whetherthey feel greatbecause the principle

69 See Dworkin,Taking Rights Seriously, supra note 63 at 358: 'Nothing could be further fromwhat I suppose than the idea thatpeople should act only in theirown interests and never in the interestsof children,lovers, friends or humanity,or thattheir votes should not represent their ideals ofjustice or other political ideals as well as their selfishinterests.' 70 The connection between votes and external preferences is subtle and important. People inevitably- and properly- vote on the basis of theirexternal preferences, says Dworkin:

[T]hey will vote for legislators,for example, who share their own theories of political justice. How else should they decide for whom to vote? But when these legislators are elected, they are subject to constraintsabout how far preference utilitarianismprovides ajustificationfor their decisions; thatis, how far the factthat a majorityprefers a particularstate of affairs(as distinctfrom the justice of what the majoritywants) counts as an argumentfor a political decision to promote it. (Ibid.) This passage needs to be read carefully.In the contextof a plebiscite, the factthat a majoritysupport a proposal makes it legitimate;and in the context of an election, the fact thata majorityshows support fora policy provides an electoral mandate for it. But in neither of these cases does it necessarilyprovide an argumentfor it (e.g., an argument that might convince a citizen to vote one way rather than the other). Whether it providesan argumentdepends on whethersomething like utilitarianism is an acceptable theoryof social policy in a given area, and it is in relation to that question that Dworkin's exclusion of external preferencesis important. To put it another way,voting doesn't purportto justify a political position (although it may, in some circumstances,be cited as evidence in a utilitarianjustification). Rather, it is a wayof choosing political positions for societyin circumstanceswhere people disagree about what is justified. But utilitarianismdoes purport to be a justificatorytheory, and in thatcontext there is somethingvery fishy about citingas part of the justificationof a principle a preference or feeling whose feltcharacter presupposes that the principle is justified.For furtherdiscussion of this distinction between first-levelarguments and second-level legitimacy,see J. Waldron, 'The Circumstancesof Integrity'(1997) 3 Leg.Theory 1 at 9-12, andJ. Waldron, Law and Disagreement(Oxford: Clarendon Press, 1999) at 195-8.

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions 392 UNIVERSITYOF TORONTO LAWJOURNAL has been vindicated or whetherthey feel offendedbecause it is violated - it almost alwaysmisrepresents their view to say that the pleasure of vindication or the pain of offence are themselves grounds for the principle in question. From the point of viewof the principle'sjustifica- tion, these sortsof pains and pleasures are mere epiphenomena. Which is not to saythat the principlemay not have a utilitarianjustification; but it is not ajustificationof thissort. I made thisdeviation into the controversyabout externalpreferences because, again, I wanted to indicate how poorlythought through Profes- sor Ellickson's conception of the harmsoccasioned bypanhandling and bench-squattingactually is. We have already seen that the annoyance aspect of being confrontedby someone who castsdoubt on one's viewof what societyis reallylike should not be counted as a harm. If anythingit is a benefit,although it is probablymore accurate to say that the simple- minded cost-benefitdichotomy is quite inadequate to deal withall this. And now we have seen that there is reason, too, for doubting whether the normative,condemnatory side of the annoyance sufferedby Ellick- son's pedestrianshould be regarded as harmfor the purposesof develop- ing anythinglike a utilitarianjustification of restrictionson the homeless. The pedestrian is irked that the bench-squatteris idle. He does not think such idleness should be permitted,so he campaigns for a law againstbench-squatting, together perhaps witha 'workfare'requirement, which would effectivelyput an end to the practice. Now, there may be good reasons in favourof thatlegal change. If the pedestrianis aware of them, then they are presumablythe reasons that weigh with him, ex- plaining whythe bench-squattingannoys him in the firstplace. But the self-awarepedestrian will not thinkthat his annoyancejustifies the proposal to change the law; instead,he willthink that the reasons thatjustify the annoyance also justifythe proposal. Suppose now that the case in favourof the workfare-and-anti-bench- squattinglaw is finelybalanced. On the one side is the arrayof reasons - - R1, R2, R3 that convinced the pedestrian; on the other side are various reasons of a humanitarianand administrativecharacter - call them Rs,Re, and R7.When the issue is finelybalanced like this,should we introduce the factthat the proposed law would relieve the pedestrian'sannoyance as an additional reason R4,operating perhaps as a tie-breaker?Surely not. The pedestrian's annoyance is not an additional factorin favourof the law over and above R1,R2,, and R3. Rather,the pedestrian's annoyance is simplythe effectthat R1,R2, and R, have in the mind of someone who believes thatthey are not being properlyattended to. This is not to say that the condemnatoryaspect of the anger that Ellickson'spedestrians feel is unimportantor not worthconsidering. It is important.But what is worthconsidering about it is the question of its

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions HOMELESSNESS AND COMMUNITY 393 justification.If Ellickson's pedestrian is irked by the fact that homeless people are violatingrules of conduct forpublic spaces, then we need to ask what these rules are and how are theyjustified; we should not be in the businessof citingthe factthat the pedestriansare irkedas a reason in support of theirviews about these rules. The External Preferenceargument may also workin another way,in the contextsEllickson considers.Many of those who are annoyed by the demands of streetpeople were broughtup to believe thatgenerosity is a virtue,and that there is somethingwrong about simplyturning one's back on a cryfor alms.71 Now, though,when the demands for help seem overwhelming,the response is oftento oppose to the feltduty of almsgiv- ing a moraljudgement to the effectthat the poor musttake responsibilityfor theirown predicament.And so one walkspast the beggars on the street,repeating in one's mind furiouslythe modem mantra 'Own fault. Own fault.' The opposition between these judgements may be experi- enced subjectivelyas anxietyand agitation,as new economic wisdom clashes with moral upbringing. Most of us feel bad aftereach of these encounters,or each of these turnings-away.But it would be wrongto cite the feelingas one of the costsof streetmisconduct. It is not an independent factorin the moral calculation; at best,it is one of the symptomsof the moral calculation's being performed.

71 For the Christianview, see Matthew25:31-46:

When the Son of man shall come in his glory,... before him shall be gathered all nations: and he shall separate them one fromanother, as a shepherd divideth his sheep fromthe goats: And he shall set the sheep on his righthand, but the goats on the left.Then shall the King say unto them on his righthand, Come, ye blessed of my Father,inherit the kingdom prepared foryou fromthe foundation of the world: For I was an hungred, and ye gave me meat: I was thirsty,and ye gave me drink:I was a stranger,and ye took me in: Naked, and ye clothed me: I was sick,and ye visitedme: I was in prison, and ye came unto me. Then shall the righteousanswer him, saying, Lord, when saw we thee an hungred, and fed thee? or thirsty,and gave thee drink? When saw we thee a stranger,and took thee in? or naked, and clothed thee? Or when saw we thee sick,or in prison,and came unto thee? And the King shall answerand say unto them,Verily I say unto you, Inasmuch as ye have done it unto one of the least of these mybrethren, ye have done it unto me. Then shall he say also unto them on the lefthand, Depart fromme, ye cursed,into everlastingfire, prepared forthe devil and his angels: For I was an hungred, and ye gave me no meat: I was thirsty,and ye gave me no drink:I was a stranger,and ye took me not in: naked, and ye clothed me not: sick, and in prison, and ye visited me not. Then shall theyalso answer him, saying, Lord, when saw we thee an hungred,or athirst,or a stranger,or naked, or sick, or in prison,and did not ministerunto thee? Then shall he answerthem, saying, Verily I say unto you, Inasmuch as ye did it not to one of the least of these, ye did it not to me. And these shall go away into everlasting punishment: but the righteous into life eternal.

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions 394 UNIVERSITYOF TORONTO LAWJOURNAL

x Complementarityagain

So I return to my little contretempswith the American Alliance for Rights and Responsibilities.The premise of their position was that 'governmentshave the right to regulate certain types of conduct in public places, to ensure that parks and sidewalksremain accessible and welcome to all.'72And as I said in section II, I agree withthis. The question is: How? On whatbasis is thisregulation to be set up? If there were no homeless persons in our communitythat is, if everyone living in our cities had access to private accommodation or guaranteed shelterspace for sleeping and for care of self, and so on - then public spaces could be regulated on the followingbasis. Since everyonewould have access to a privatehome, activitiesdeemed particu- larlyappropriate to the privaterealm - activitieslike sleeping, copulat- ing, washing,urinating, and so forth- could be confined to that realm. Public places could be put off-limitsto such activities,and dedicated instead to activitiesthat complementedthose that citizens performedin their own homes. (This is the ComplementarityThesis I mentioned at the beginningof thisessay.)"7 Public places could be dedicated to things like strolling,picnics, meeting people, walkingdogs, children'splay, and so on. It would be reasonable for those who wanted to enjoy the public spaces to expect not to find people sleeping, cooking, or storingtheir possessions there, and not to find evidence of human urination or defecation. They could reasonablyassume thateveryone had a home to go to foractivities of thatkind. Laws and regulationsprohibiting sleeping and storingpossessions on sidewalksand public parkscould be enforced in the lightof thatexpectation without fear of disparateimpact. That, in myview, captures the spiritof the AARRbrief, which (as I said earlier) evoked a timewhen citizensfrom all walksof life- ' [t]hose with Armani suits,and those withnose rings;elderly people and gay couples; residentsand visitors;rich, middle, and strugglingclasses'74 - spent their leisure hours in public places.75It is an attractiveidea, of public places facilitatinginteraction among strangers,not just among friends,76and as such it is preferableto the usual nostalgiafor Gemeinschaftthat one finds in the communitarianliterature.77 As AARRcounsel Rob Teir puts it,

72 AARRbrief inJoyce, supra note 7 at 1. 73 Towardsthe end ofsection II, supra. 74 Teir,'Restoring Order,' supra note 1 at 290. 75 AARRbrief in Joyce, supra note 7 at 2. 76 See also the discussionin M. Ignatieff,The Needs of Strangers (London: Chatto& Windus,1984) at 139ff. 77 See alsoJ. Waldron, 'Particular Values and CriticalMorality' 77 Cal.L.Rev.561 at 581 ff.,reprinted in LiberalRights, supra note 5, 168 at 193ff.

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions HOMELESSNESS AND COMMUNITY 395

Cityparks and sidewalkswere built to be communitymeeting places, where people of differentraces, religions, ethnic groups, socio-economic levels, and politicalviews, could come togetherand sharein thebenefits of public spaces. These venuesare places of integration,assimilation, mixture of social classes, and a counterweightto theincreasing fragmentation ofsociety.78 We are to imagine diverse citizens coming out into the parks and boulevards where theycan enjoy theirleisure, 'interact with their fellow citizens and leave behind their isolation and segregation'79before returningonce again to the privaterealm.Certainly this is an attractive image. Unfortunately,it is not appropriate for the regulationof public places in a societywhere thereare large numbersof homeless people. In such a society,public spaces have to be regulated on a somewhatdiffer- ent basis. They have to be regulatedin lightof the recognitionthat some people have no private space - not even the temporaryprivacy that public shelters or public toiletswould afford- to come out of or to returnto. Fairness demands that public spaces be regulated in light of the recognitionthat large numbersof people have no alternativebut to be and remain and live all theirlives in public. For such persons,there is an unavoidable failure of the complementaritybetween the use of privatespace and the use of public space, and unless we are prepared to embrace the mostegregious unfairness in theway our communitypolices itselfin public, we are simplynot in a position to use thatcomplementar- ityas a basis forregulation." It is worth dwelling on the question of fairness.Professor Ellickson recognizes it as a formidableissue,"s but the AARR'scounsel Robert Teir proteststhat

78 'Maintaining Safetyand Civility,'supra note 1 at 289. 79 Ibid. at 337. 80 Whetherthis works as a constitutionalargument is another matter:one has to contrast the argument that worked in Pottingerv. City of Miami, 810 F.Supp. 1551 at 1571 (S.D.Fla. 1992) - holding that the 'practice of arrestinghomeless individuals for performinginoffensive conduct in public when theyhave no place to go is cruel and unusual in violation of the eighth amendment, is overbroad to the extent that it reaches innocent acts in violation of the due process clause of the fourteenth amendment and infringeson the fundamentalright to travelin violationof the equal protection clause of the fourteenthamendment' (ibid. at 1583) - withthe argument thatwas rejected in Joycev. Cityof San Francisco,846 F.Supp. 843 (N.D.Calif., 1994). But even assuming thata constitutionalchallenge fails,the factthat the legislation is constitutionaldoes not mean it is fair.This essay should be read as a consideration of the issue of fairnesswithout much regard to the political forumin which thatissue mightbe explored. 81 See 'Controlling Chronic Misconduct,' supra note 12 at 1247: 'The reconciliation of individual rights and communityvalues on the streets is a profoundly difficult problem.'

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions 396 UNIVERSITYOF TORONTO LAWJOURNAL

[t]here is nothingunfair or mean-spiritedabout wanting to be freefrom harass- mentand intimidation,wanting urban parks where children can playand adults can enjoythe green, and thequiet, or wanting urban parks that are notfilled with litter,human waste, needles, bedrolls, drug users, and used condoms....These rules... are setso thatall peoplefeel welcome in the public spaces.82 He is right.In itself,the aspirationhe mentionsis just and admirable. Unfairnesscomes into the pictureonly when we consider the implemen- tation of this goal against the background of homelessness. But again Teir protests.He says that those who sufferfrom the deteriorationof public places are oftenthe poorest,rather than the most prosperous,of those who have a home to go to; this is not a case of discrimination against the poor in favourof the rich;at worst,it is discriminationagainst one class of poor people in favourof another.8"And anyway,he says,the measures discussed in thisarticle are not aimed at the homeless. Rather, they are aimed at and address conduct,and only those who choose to engage in the prohibitedconduct fallwithin their reach.84 Once again, he maybe right- at least as to the intention.85But one is entitled to- consider, also, evident and predictable disparities that

82 'RestoringOrder,' supra note 1 at 290. 83 Thus Teir writes,ibid. at 290-1

Nor are these measure unfair to the poor.... [I]t is not the affluentwho reap the benefitsof these measures. The rich,after all, can take care of themselves.They are not,speaking generally,dependent upon public parksfor recreation. They usuallylive in secured communities,and shop in safe and comfortableplaces. The well-offcan also leave an area when it gets intolerable. Rather,it is the poor and middle-classes who depend upon the safetyand civilityof public spaces. They have feweroptions about relocating, less options about schools, and less options about private recreational places. Ellickson takes a similar line in 'Controlling Chronic Misconduct,' supra note 13 at 1189-90:

Most beggars and bench squatters are economically and socially destitute. For observersconcerned primarilywithdistributivejustice, extreme poverty might furnish ... a sufficientreason forsiding witha disorderlystreet person in any policy context. This is an ill-considered position. To favorthe poorest may disadvantage the poor, who are as unhappy with street disorder as the rest of the population. Because residentsof poor urban neighborhoods tend to make especially heavyuse of streets and sidewalksfor social interactions,they have an unusuallylarge stakein preventing misconduct there.

These are perfectlyreasonable points.The issues about fairnessdiscussed in the text concern, not rich/poor comparisons,but the enormous differencein burden cast by public-spaceregulations on those who are homeless and those (whetherrich or poor) who are not. 84 Teir, 'RestoringOrder,' supra note 1 at 291. 85 For some doubts, see Waldron, 'Homelessness,' supra note5 at 313-4 (LiberalRights at 327-8), settingout four reasons for doubting the impartialityof the intentionsof those who lobby for prohibitionson sleeping in public.

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions HOMELESSNESS AND COMMUNITY 397 differentclasses of people will experience in bringing their behaviour withinthe norms he wants to enforce. If not as a constitutionalmatter, then certainlyas a matterof justice, those who have the power to regulate public places must pay special attentionto the differencebetween the impact of a given regulation on a person who has a home and its impact on someone who is homeless. In the case of a person who has a home, compliance withan ordinance prohibiting,for example, sleeping in public places is simplya matterof relocation.For someone who has no home, however,and - let's say- no access to a shelter,86compliance with such an ordinance would mean that he mustnot sleep (for there is now no place where his sleeping is permissible).87This may not be what any enforceror communitarianlobbyist intends, but it is the easily foresee- able result of a number of familiar intentional prohibitions.88This impact is so qualitativelydifferent from the impact of the regulationon the person who has a home to returnto that it amounts almost to the application of a quite differentset of laws. Certainly,the ordinance, if enforced, would have an impact on the homeless whose cruelty (the denial of sleep, period)was out of all proportionto the minor inconve- nience that would be sufferedby a person who had somewhere else to sleep. And, of course, eventually,the effectiveprohibition on sleep becomes physiologicallyimpossible to comply with.81'(Against all this,

86 More about thiscondition in a moment: see textaccompanying note 93 infra. 87 Waldron, 'Homelessness,' supra note 6 at 315 (LiberalRights at 328-9): For a person who has no home, and has no expectation of being allowed into something like a privateoffice building or a restaurant,prohibitions on thingslike sleeping thatapply particularlyto public places pose a special problem. For although thereis no general prohibitionon acts of these types,still they are effectivelyruled out altogether for anyone who is homeless and who has no shelter to go to. The prohibitionis comprehensive in effectbecause of the cumulation,in the case of the homeless, of a number of differentbans, differentlyimposed. The rules of property prohibitthe homeless person fromdoing anyof these acts in private,since thereis no privateplace thathe has a rightto be. And the rules governingpublic places prohibit him from doing any of these acts in public, since that is how we have decided to regulate the use of public places.... Since privateplaces and public places between them exhaust all the places thatthere are, there is nowhere thatthese actions may be performedby the homeless person. And since freedom to performa concrete action requires freedomto performit at some place, itfollows that the homeless person does not have the freedomto performthem. If sleeping is prohibitedin public places, then sleeping is comprehensivelyprohibited to the homeless. 88 For the detailed argument here, see ibid. at 315-7 (LiberalRights at 329-32). 89 Ibid. at 320 (LiberalRights at 334-5): '[A]ny restrictionon the performanceof these basic acts has the featureof being not only uncomfortableand degrading,but more or less literallyunbearable forthe people concerned. People need sleep, forexample, not just in the sense thatsleep is necessaryfor health, but also in the sense that they will eventuallyfall asleep or drop fromexhaustion ifit is denied them. People simply cannot bear a lack of sleep, and theywill do themselvesa great deal of damage trying to bear it. The same, obviously, is true of bodily functions like urinating and

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions 398 UNIVERSITY OF TORONTO LAWJOURNAL

Ellickson's observationthat '[w] hile no one's willis fullyfree, virtually all of us have some capacityfor self-control'does seem a littlelame, to say the least.) '0 I said a moment ago thatwe should assume for the sake of argument that the homeless have no publiclyprovided shelter to go to. For manyof the homeless thisis false,91although it is trueduring daylight hours, and importantlytrue then so far as publiclyprovided bathrooms are concerned. I suggested this assumption - artificialas it is - in order to emphasize the extent to which the fairnessof public-place regulations depends on other aspects of provision for the homeless. Robert Teir believes that the two issues are separable. He cites a commentby colum- nistGeorge Will to the effectthat '[t] he question of whatsociety owes in compassionate help to streetpeople is, surely,severable fromthe ques-

defecating.These are thingsthat people simplyhave to do; any attemptvoluntarily to refrainfrom doing them is at once painful,dangerous, and finallyimpossible.' This - as I said in the 'Homelessness' article (ibid. at 320; Liberal Rightsat 334) - is somethingthat every torturer knows: '[T]o break the human spirit,focus the mind of the victimthrough pettyrestrictions pitilessly imposed on the banal necessities of human life.We should be ashamed thatwehave allowed our lawsof public and private propertyto reduce a million or more citizensto somethingapproaching thislevel of degradation. Increasingly,in the waywe organize common property,we have done all we can to prevent people fromtaking care of these elementaryneeds themselves, quietly,with dignity, as ordinaryhuman beings.' 90 Ellickson, 'ControllingChronic Misconduct,'supra note 6 at 1187. The fulltext of the passage reads (footnotesomitted): Many advocates sincerelybelieve thatstreet people are so constrained by economic and social circumstancesthat theylack real choices. Most (although not all) social- welfare professionals hold the view that poor people always act under duress; according to thisview, society should not 'blame' poor people or, under an extreme formulation,ask them to bear any responsibilities.While no one's will is fullyfree, virtuallyall of us have some capacity for self-control.Legal and ethical systems thereforeproperly subscribe to the proposition- or salutarymyth - thatan individual is generallyresponsible forhis behavior. This policy,at the margin,helps fostercivic rectitude.To treatthe destituteas choiceless underestimatestheir capacities and, by failingto regard them as ordinarypeople, risksdenying them full humanity. 91 For example, New York Cityguarantees a bed in a shelter to any person who seeks it (subject to some workand conduct requirements):see P.T. Kilborn, 'Gimme Shelter: Same Song, New Tune' The New YorkTimes (5 December 1999) 5. But New York is exceptional among American cities in this regard. (See also Donna Wilson Kirchheimer,'Sheltering the Homeless in New York City:Expansion in an Era of GovernmentContraction' (1989) 104 Pol.Sci.Quar. 607.) A reportcommissioned by the United States Conference of Mayorsbetween 1986 and 1995 in twenty-ninemajor cities indicates thatin 1995, forexample, 24% ofshelter requests went unmet during that year and that emergency shelters in 79% of the surveyed cities turn away homeless familiesbecause ofa lack of resources.See L.D. Waxman, K. Peterson, & M. McClure, U.S. Conferenceof Mayors,a Status Reporton Hunger and Homelessnessin America'sCities: 1995 (1995) at 59-60, cited in C.V. Tusan, 'Homeless Families from 1980-1996: Casualties of Declining Support for the War on Poverty' (1997) 70 So.Calif.L.Rev. 1141 at 1144.

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions HOMELESSNESS AND COMMUNITY 399 tion of what right the communityhas to protect a minimallycivilized ambience in public spaces.'"2 The analysisI have given shows why the questions are notseverable: the less the societyprovides in the way of public assistance,the more unfairis itsenforcement of norms forpublic places that depend on a complementaritythat simply doesn't apply to a considerable numberof citizens.Notice, though,that this does not mean that all regulationof public places mustbe suspended untilprovision has been made for the needs of the homeless. From what I have said here, nothingfollows about the enforcementin public places of those rules of conduct thatapply everywhere:murder and assaultwill remain illegal in public as well as in private,and ifthe use and supplyof narcoticscontin- ues to be banned, then theytoo will be banned in public places. The argumentsI have made primarilyconcern necessaryactivities - such as sleeping and urinating- whichwould be perfectlylegal if performedin private.So far as these actions are concerned, it is quite unfair to ban them in public - to say to the homeless person in a public street or square, 'This is not the place for thatactivity' - ifwe have not at the same time taken steps to ensure thateveryone has access to a privateplace in which to performthem. So long as we have not taken those steps, then banning such activitiesin public is like enforcing a curfewagainst the homeless when there is nowhere for them to go. It fails to take proper account of the factthat, as thingsstand, public places are the onlyplaces theyare allowed to be. If, as a result, the visibilityof these properlyprivate (but, for the homeless,unavoidably public) activitiesis a deterrentto more traditional uses of public space by citizenswho do have a home to go, then we have no choice but to say to the latter,'In the presentstate of public policy,we can longer guarantee the use of public space forthis sort of respectable activityalone. Anyuse you make of the public space willjust have to put up with the company of hundreds of people washing, urinating,and sleeping in the area where you want to have your demonstration,or do yourjuggling, or play pick-upfootball, or lay out your picnic, or make your promenade.' Now, this is certainly a matter of regret,and the consequence maywell be, as Ellicksonhas suggested,an impoverishment of the public dimension of culture and civilsociety, as those who have a choice flee the downtownstreets and parks and take refuge in cyberspace, suburbanmalls, or gated communities,leaving public places to the mercyof those who have no option about remaining there.93But it is importantto see thatthis is not the sortof dilemma thatwe can solve by simplyadjusting the regulations.We can't have it all ways:unless we are willingto take the next step and actuallyeradicate the homeless or put

92 G.F. Will, 'Beggars and Judicial Imperialism' The WashingtonPost (1 February 1990) A21, cited in Teir, 'Maintaining Safetyand Civility,'supra note 1 at 337. 93 Ellickson, 'Controlling Chronic Misconduct,' supra note 12 at 1172.

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions 400 UNIVERSITY OF TORONTO LAWJOURNAL

them out of sightin concentrationcamps,'94 then theyand theirlives are just there- unavoidably- byvirtue of the factthat distribution of housing has leftthem nowhere else to go. And our use and regulationof public places mustbe adjusted accordingly. I suspectthat an apprehension of thiscollapse of the complementarity on which the traditionaluse of public space depended lies at the heart of many of the attitudesthat Robert Ellickson discerns in his pedestrians. The rules that concern them - the rules theyare distressedto see the homeless flouting- are predicated on the traditional public/private complementarity;and the distressthey feel may reflectan angryfrustra- tion arisingfrom their knowledge that the basis forthe fairenforcement of those rules no longer exists.If thisis true,then there is an additional mistake in Ellickson's citingsuch frustrationas a utilitarianground for persistingwith the rules in question, or enhancing or intensifyingtheir enforcement.For now we see that the verydistress he regardsas a harm of the homeless people's activityis in factfrustration associated withan apprehension of the unfairnessof effectivelyprohibiting that activity, indeed the moral impossibilityof the whole traditionalframework of rules of that kind. Once again we see how badly one can go wrong by simplycategorizing everynegative emotion thatpeople feel as a costfor the purpose of an economic calculus.

XI Social norms

Let me head for the finishon a more conciliatorynote. The best work thatEllickson has done concerns the emergence and role of social norms - informalcustoms and practices addressing problems and conflicton the ground, so to speak, to which formal state or sovereign law often takesa distantsecond place."9In his 1996 articleon 'Chronic Misconduct in Public Places,' Ellickson recognized thatthis social normsperspective

94 For apprehensions along these lines see, e.g.,Bob Pool, 'Fanfare,Fear Surround New Shelter' Los AngelesTimes (16 April 1999) at Bl: Development of the centerwas firstproposed in 1994 by downtownbusiness leaders and supported byMayor Richard Riordan. Its original concept called fora $4 million urban campground servingas manyas 800 homeless people on a fenced-inlot. As part of that plan, outreach vans would circulate through downtown streets and social servicesworkers would invitetransients to ride withthem to the center. But the city scaled back its proposal after critics such as the Los Angeles Coalition to End Homelessness blasted the plan as 'a first step on a slippery slope down to concentration camps in rural areas for homeless people.' 95 See especially OrderWithout Law, supra note 13, discussinginformal rules about cattle trespass in Shasta County,California. See also R.C. Ellickson, 'Of Coase and Cattle: Dispute Resolution among Neighbors in Shasta County' (1986) 38 StanfordL.Rev. 623 and 'Law and Economics Discovers Social Norms' (1998) 27J.Legal Stud. 537.

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions HOMELESSNESS AND COMMUNITY 401

ought to apply to citystreets also. He invoked an oft-citeddictum from JaneJacobs's classic studyof urban organization:

[T]he public peace - the sidewalkand streetpeace - of citiesis not kept primarilyby the police, necessaryas police are. It is kept primarilyby an intricate,almost unconscious, network of voluntarycontrols and standards amongthe people themselves,and enforcedby the people themselves....No amountof police can enforcecivilization where the normal, casual enforcement ofit has brokendown.96

He also expressedsome apprehension about the obstacles thatdefeat or undermine thispattern of 'normal, casual enforcement,'so faras the prevention of chronic annoyance by the homeless is concerned. In principle, it should be possible, he said, 'fora bystanderto interveneto preventa streetperson fromannoying another sidewalkuser.'97 Even a person who is 'reluctantto chastisea panhandler maybe willingto frown at an almsgiver.'98But, he laments, these thingsseldom happen. His explanation has to do mainly Withthe public goods aspect of street order, especially in a large metropolitan city,where encounters are anonymousand not necessarilyrepetitive as betweenthe same 'players.'99 (Here we come back to the point about the insignificanceof harms inflictedby these annoyances on any one occasion.)100 But there's also another explanation,which may be more charitable to Ellickson's pedestrians.We have seen that because of the failure of complementarity,the conditions for the legitimacy and fairness of traditionalnorms of conduct in public places now no longer apply,so far as a significantnumber of citizensare concerned. I suspect that this is widelyrecognized among Ellickson's pedestrians,and thatit accounts in part for theirfailure to participateenthusiastically in the informalsocial enforcementexercise thathe recommends.True - as I argued earlier101 - that recognitionof the failureof complementarityis not particularly cheerful. Often it is associated with anger, best explained in termsof cognitive dissonance and denial. This leaves many people acutely uncomfortableas theystruggle to express,regulate, and make sense of the annoyance that accrues frombeing confrontedwith the spectacle

96 J.Jacobs, TheDeathand Lifeof Great American Cities (NewYork: Random House, 1961) at 31-2, quoted by Ellickson in 'Controlling Chronic Misconduct,' supra note 12 at 1196. See also R.H. Pildes, 'The Destructionof Social Capital Through Law' (1996) 144 U.Pa.L.Rev. 2055 at 2061 ff.and Livingston,supra note 54 at 558. 97 'Controlling Chronic Misconduct,' supra note 12 at 1196. 98 Ibid. at 1197. 99 Ibid. at 1196-7. 100 See supra section IV, especially notes 31-2 and accompanying text. 101 See supra section VII, especiallynote 50 and accompanying text.

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions 402 UNIVERSITY OF TORONTO LAWJOURNAL and the demands of streetpeople. But the pedestriansthat I see seem more aware than Ellickson appears to be of the conditional nature of streetregulation. The regulationof public places is not like the categori- cal imperative;it is sensitiveto conditions and circumstances.I think most people recognize thatwith the advent of large-scale homelessness in our cities,conditions, and circumstanceshave changed; and, while this maynot necessarilyexcite a greatdeal of politicallyeffective compassion, still its implicitrecognition may be enough to undermine - and (in an indirectand no doubt back-handedWay) to appropriatelyundermine - the sense of there being any acceptable ethical basis forthe ordinarycitizen to chastise a panhandler or to roust and admonish a person who has fallen asleep on the street. I don't mean that normativityhas disappeared altogetherfrom the streets.Quite the contrary:new norms may be emerging,or new selec- tionsbeing made in new circumstancesfrom among the traditionalsocial rules. For example: manywho give regularlyto panhandlers understand the importance of norms of time,place, and manner so far as begging for change is concerned, and they do what theycan to enforce these - even if only by selective giving. (I know from conversationthat many people willcontemplate giving only in circumstanceswhere all partiesto the transactionhave an opportunityto move awayif theywish; theywill not, for example, under any circumstances,give money to a panhandler in a crowded subwaycar.)102 Also, the more or less permanent presence of the homeless may actuallyenhance streetorder in certain ways.Jane Jacobs spoke of the importance of 'eyes upon the street';she mentioned the special incen- tivesof merchantsand regularusers to observeand controlwhat goes on in a particular locale.10' Regular panhandlers have their eyes on the street day and night,and oftenthey know it as well as or better than its more prosperous users.They oftenform bonds of affability,trust, even - paradoxically- protectivenessvis-A-vis their regular alms-giving clientele. And their sense of what is going on may well equal or even surpass - occasionallythey may be called on to supplement- the streetknowledge of an alert patrol officer.104

102 Ellickson's proposed testfor this - 'How oftenhas a New York commuter countered a subway panhandler's monologue by, for example, startinga chant of "Justsay 'no'"'?' ('Controlling Chronic Misconduct,' supra note 12 at 1197) - is arguably somewhat under-inclusive. 103 Jacobs, supra note 98 at 35-7, mentioned by Ellickson in 'Controlling Chronic Misconduct,' supra note 12 at 1197-8. 104 However, Ellickson deprecates this. Responding to a suggestion in B.J. Goldstein, 'Panhandlers atYale: A Case Studyin the Limitsof Law' (1993) 27 Ind. L.R. 295 at 346, that 'unaggressive chronic panhandlers may aid in police effortsto curb boisterous transients,'Ellickson writes ('Controlling Chronic Misconduct,' supra note 12 at

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions HOMELESSNESS AND COMMUNITY 403

As Ellickson well knows, the emergence and sustenance of social norms is not a simple or predictable business. Such norms will not emerge in response to everysituation about which the indignantcitizen thinks,'Something ought to be done.' The closeness to the ground (so to speak) of the conditions under which social norms are constructed means thatthey are likelyto be much more sensitive(than, say,a legisla- ture is) to the detailed predicaments of various persons in a given situation and to others' awareness of those predicaments.If a norm is manifestlyunreasonable at the human level, then it may prove unen- forceable, since its enforcementrelies on the voluntaryintervention of those who are closelyacquainted withthe basis of the unreasonableness. The contrastwith the enforcementof officialstate law is importanthere. In the case of a social norm,extra enforcementresources that are blind to, or unaffected by, the sense of the norm's unreasonableness can simplybe wheeled in to supplant the reluctance of the enforcerson the ground. (Social normscannot mobilize taskforces.) So, forexample, ifit is well known that there are no public restroomsin a givenvicinity, and that homeless people have access to shelterfacilities only at night,then social normsare unlikelyto springup enforcingabsolute prohibitionson public urination.Instead, such social normsas thereare on thisissue will likelydirect street people to particularlocations (in parks and public gardens,etc.) where a modicum of privacyis available and where offence to other members of the public is less likely.105And the street-level reasonableness of social normsin this regard is likelyto be echoed also, to a certain extent, in the exercise of ordinarydiscretion by police officerson patrol. Someone urinatinginto the fountainin Lincoln Plaza may be arrested;someone who has been seen behind a bush in Central Park maynot be. These conclusionsabout social normsare significantfor an evaluation of the various legislativeproposals that Ellickson and also the AARRput forward.In Ellickson's presentationparticularly, the proposals are put forwardas a legal response to the failureof social norms:it is the familiar idea of law steppinginto the breach where externalitiesand transaction costs have defeated more informalprocesses. In fact,however, the im- plications of the failureof social normsin thisarea may be more direct:

1198n), 'Most chronicpanhandling, however, occurs in well-traffickedlocations, where thereare likelyto be numerousother eyes upon the street.A chronic panhandleris thereforeunlikely to makea netcontribution to streetorder.' 105 ProfessorEllickson and I wereonce givena tourof such sites in downtownAtlanta by a homelessperson (during a breakin theconference mentioned supra note t). Itwas madeclear to us that,among street people, the designation of such sites was normative as wellas matter-of-fact- these were known as 'appropriate'places to, for example, urinate- althoughI shouldadd thatI don'tremember the word'normativity' actually beingused.

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions 404 UNIVERSITYOF TORONTO LAWJOURNAL theirlack ofviability may provide useful information about the unreasonableness of the proposed scheme of regulation by demonstrating,in effect,that those who would have to enforce them do not have the stomach to do so againstpeople who have littlechoice but to be in some sense in violationof such norms.

XII Conclusion

The problem of homelessnesshas the potentialto show communitarian- ism at itsbest and at itsworst. At its best, communitarianismrefuses to be browbeaten by either conservativeor liberal theoriesabout what a good societymust be like: it encourages us to look to the conditionsof viability of actualcommunities, and it predicts that those conditions may sometimessurprise us, which means thatactually existing communities have the potential to teach us something about social and political ideals. 'A community,'in Philip Selznick's definition,'is a comprehensiveframework for social life.'106It provides a basis on which various human concerns are felt, values pursued,and commitmentsentered into,among relativelylarge numbers of people who accept that,since theyare livingmore or less permanently in one another's company, they have a responsibilityto orient their actions and relations to the shared environment- moral as well as physical- in which all of them must live. 07 In this respect. there is a certain sociological given-nessabout community:it contrasts,on the one hand, with utopia and social ideal, and, on the other hand, with club, cult, and commune. One finds oneselfamember of a given community, not by enlistinga group of like-mindedchums, but by understanding who it is thatone is 'unavoidablyside-by-side' with, and who it is that the impact of one's actions and decisions is going to be feltby, whether one chooses this or not. The fellow members of one's communityare not necessarilypeople like oneself: theyare, rather,those with whom 'one cannot avoid interacting.'108One is already in a community,already in communitywith specified others; and one's communitarianresponsibili- ties springfrom that, not fromone's wishesor ideals. Thus the fact that someone smells bad, looks dishevelled,or is not the person one would choose to associate withdoes not mean thathe is not a member of one's community.If he is there,on the streets- the verystreets that are the basis of one's social, commercial, recreational interactions- then he is a

106 P. Selznick, 'The Idea of a CommunitarianMorality' (1987) 75 Cal.L.Rev. 445 at 449. 107 See P. Selznick, The Moral Commonwealth(Berkeley: Universityof California Press, 1992) at 357-71. 108 These formulationsare drawnfrom I. Kant, TheMetaphysics ofMorals, trans. M. Gregor (Cambridge: Cambridge UniversityPress, 1991) at 121-4.

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions HOMELESSNESS AND COMMUNITY 405 member of the communitytoo. And any storyone tellsabout communal rightsand responsibilitiesmust take him and his interestsinto account. But there is also a more disturbingside to contemporarycom- munitarianism.Communitarians often seem to be yearningfor forms and images of communitythat are no longer present or reallypossible. They conjure up a delightfulimage of community,a sort of Norman Rockwellpicture of cheerful,prosperous people interactingrespectfully on the streets,in the parks,at church,in townmeetings, at LittleLeague games, looking out forone another,and sustainingcherished traditions of civility,participation, civic boosterism,and mutual aid.1'9 Nothing is too crowded,no one is ever scared or intimidated,nothing or no one is dirtyor in despair,and thereare no broken windows.I guess thatas long as these yearningsare confined to the realtor'sbrochures for actual or notionally'gated' communities,or to variousDisneyfied experiments in urban reconstruction,110they are probably no more harmfulthan any other cosmetic embodiment of American schwarmerei.But theybecome dangerous, and they portend great injustice,when they are used as a basis fordetermining rules of social conduct in communitysettings that do not initiallymeet their specifications.When this happens, then, although the resultingrules maybe called'community norms,' what they

109 See the extractfrom the AARR brief in Joyce,quoted at text accompanying note 9 supra. 110 The most explicitsuch experimentis the town of Celebration, Florida. See D. Young, 'The Laws of Community:The NormativeImplications of Crime, Common Interest Developments, and "Celebration"' (1998) 9 Hast.W.L.J.121: Celebration representsthe privatizationand increased corporate control of whole aspects of the American lifestyle,a bastardizationof the term'community.' Much like common interestdevelopments, Celebration functionslargely by conformityand control ratherthan truecultural consensus and communitydecision-making. Though Celebration will have many of the semblances of a real town, for example, 'The Celebration Town Hall,' 'The Workplace,' and 'The Institute,'there is no real town government.The closest thingto representationin Celebration is membershipin the homeowner association. However,even theactions of thishomeowner association can be unilaterallyoverruled by the corporation. The sense of communityso often yearned for seems to be ironicallymisguided in the 'prettycalm' and 'civic infancy' of Celebration. The longing for a bygone America, 'is a yearningfor civic maturity' in which the 'messy responsibilityof democracy held sway,and society worked.' Celebration does not offerthese attributes.

See also A. Ross, TheCelebration Chronicles: Life, Liberty and ThePursuit of Property Values in Disney'sNew Towns(New York: Ballantine, 1999); M. Pollan, 'Town-BuildingIs No MickeyMouse Operation' TheNew YorkTimes (14 December 1997) at 56; D. Tarrant, 'No Cause for Celebration: Disney-createdTown Fell Short of Being the Utopia Its Planners Envisioned, but It's Not the Set for Another Truman Show, Either' Sun- Sentinel[Fort Lauderdale] (22 August 1999) ID; andJ.V. Iovine, 'A Tale of Two Main Streets:The Towns That Inspired Disney Are Searching for a Little Magic of Their Own' TheNew YorkTimes (15 October 1998) Fl.

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions 406 UNIVERSITYOF TORONTO LAWJOURNAL actuallyrepresent is the impositionof a particulara priorivision by a few persons (often outsiders) on a communitythat has traditionallyorga- nized itselfon quite another basis. A paragraph or twoago,"' I said thatan attractivecommunitarianism takes communitiesand theirmembers as it findsthem, not as its propo- nentswish they would be. Those who have responsibilityfor the health of the common environmentin a particularplace are those who actually inhabit thatplace. This principleof the given-nessof communityis quite rightlyinvoked by Ellickson, Teir, and otherswhen theyargue thatstreet people too have responsibilitiesto the community- responsibilities,for example, for the condition and safetyof the community'spublic spaces. Whether or not a homeless person has any choice about being on the streets,the sheer factof his being there means that he too has a dutyto the communityin that regard. This we can accept. What we cannot accept, however, is that the definition of communal responsibilities should proceed on a basis thattakes no account of the predicamentof the homeless person or of the particularnature of the stake that she may have in the way public spaces are regulated. If the norms for public spaces are to be observed byhim, then the logic of genuine as opposed to cosmetic communitarianismrequires that those normsbe constructedin partforhimas well.We are not entitledto insistthat the homeless person abide bycommunity norms, or thatthose normsbe enforcedagainst her, if the norms are constructedin an image of communitywhose logic denies in effectthat homelessness exists. In the end it comes down to a connection between communityand authenticity.In myparticular criticisms of Ellickson,in sectionsVII and Ix of this essay, I wanted to insist that the calculus of social costs and benefitsshould not be used in a waythat panders to the denial and the fantasizingthat distorts much Americanthinking about injustice.But the broader message about homelessness and communityas such is the following.So long as people live among us in a condition of homelessness,our normativedefinitions of communitymust be responsiveto their predicament; and it must be responsive,not only in articulatingsome vague sense of social obligation to 'do something' about the problem, but in accepting that the verydefinition of communitymust accommo- date the stake thatthe homeless have - as communitymembers - in the regulationof public places. If the call fora greateremphasis on 'communityvalues' helps us see that,and helps us see our way throughto new and more hospitable conceptions of communal responsibility,then well and good. But, as thingsstand, the call is mostoften heard in connection withschemes of regulationthat simplytry to wish homeless members of the communityaway. So long as thatis the case, the moral credibilityof modern communitarianismremains a matterof doubt.

111 See textaccompanying note 110 supra.

This content downloaded from 130.63.180.147 on Mon, 6 Jan 2014 13:51:41 PM All use subject to JSTOR Terms and Conditions