Homelessness and Community

Homelessness and Community
Author(s): Jeremy Waldron

Source: The University of Toronto Law Journal , Autumn, 2000, Vol. 50, No. 4
(Autumn, 2000), pp. 371-406
Published by: University of Toronto Press
Stable URL: https://www.jstor.org/stable/825960
REFERENCES
Linked references are available on JSTOR for this article:
https://www.jstor.org/stable/825960?seq=1&cid=pdf-
reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.
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].
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
University of Toronto Press is collaborating with JSTOR to digitize, preserve and extend access
to The University of Toronto Law Journal
This content downloaded from

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
All use subject to https://about.jstor.org/terms
Jeremy Waldron* HOMELESSNESS AND COMMUNITYt
I Introduction
I am not a communitarian, but I was prompted to ask questions about

the relation between homelessness and community by an encounter in
California some years ago with an organization called The American Alli-
ance for Rights and Responsibilities (AARR). The AARR is an activist arm of
The Communitarian Network,' litigating to defend local and municipal
initiatives that promote what it regards as increased safety, civility, and
* Maurice and Hilda Friedman Professor of Law and Director of the Center for Law
and Philosophy, Columbia University.
t This is a slightly modified version of the Cecil A. Wright Memorial Lecture
delivered at the University of Toronto Law School in February 2000. Earlier version
were presented, under the title Homelessness, Community and Denial, at a 1998 Con
ference on Homelessness at Georgia State University in Atlanta and at a meeting of
Columbia University's Fifteen Minute Paper Group. I am grateful to all the
participants on those occasions - particularly Robert Ellickson - for their comment
and suggestions.
1 D.B. Kopel & C.C. Little, 'Communitarians, Neorepublicans, and Guns: Assessin
the Case For Firearms Prohibition' (1997) 56 Mar.L.Rev 438 at 443: 'The
Communitarian Network has created an activist arm to implement its ideas on a
grassroots level: the American Alliance for Rights and Responsibilities.' The AARR is
a non-profit public interest group founded in 1988 and described by one of its
leading litigators as aiming to 'restore the spirit of community in the United
States.... [Its] approach is to identify, promote and defend new approaches which
make citizens part of the solution, and strike a balance between extreme rights
claims and those who would sacrifice civil liberties as means to an end.' R.S. Golden,
'Toward a Model of Community Representation for Legal Assistance Lawyering:
Examining the Role of Legal Assistance Agencies in Drug-related Evictions from
Public Housing' (1988) 17 Yale L.& Pol'y Rev. 527 at 552n, quoting Affidavit of
Roger Conner in Support of Motion To Modify and Intervene, in Escalera v. NYCHA,
924 F. Supp. 1323 (S.D.N.Y. 1996). In recent years, the AARR has taken on a new
name, 'The Center for the Community Interest'; 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 following articles by Robert Teir, leading counsel for
AARR, provide a good account of the organization's mission: 'Restoring Order in
Urban Public Spaces' (1998) 2 Tex.Rev.L. & Pol. 255; 'Maintaining Safety and Civility
in Public Spaces: A Constitutional Approach to Aggressive Begging' (1993) 54
Lou.L.Rev. 285 [hereinafter 'Maintaining Safety and Civility].
(2000), 50 UNIVERSITY OF TORONTO LAWJOURNAL

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
372 UNIVERSITY OF TORONTO LAWJOURNAL
community responsibility.2 The AARR filed an amicus brief supporting the

City of San Francisco in some litigation brought by homeless advocates
against former Mayor Frank Jordan's 'Matrix' program,3 which was in-
tended as a coordinated set of initiatives by police and social services to
remove encampments of homeless people from the centre of San
Francisco (particularly the 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 homelessness that I pub-
lished in 1991,5 he asked if I would be willing to file a response to the
AARR brief. Needless to say, the lawsuit was unsuccessful; actually it was
eventually declared moot, the City of San Francisco having abandoned
the program in question after the election of Mayor Willie Brown in
1995. But the issues that were raised in the exchange I had with the
American Alliance for Rights and Responsibilities remained with me, and
I want to talk about some of those in this essay.
II Community control of public places
The line taken by the AARR was a familiar one, and it is one they have
pursued in a number of cities in the United States.6 Mayor Jordan's
2 Cases with AARRintervention have included Asquithv. City ofBeaufort, 139 F.3d (1998),
supporting South Carolina municipal noise ordinance against challenge by street
preachers; Johnson v. Rodriguez, 110 F.3d 299 (1997), supporting consideration by
Texas parole boards of letters from members of public against prisoner release; Doe
v. Pataki, 120 F.3d 1263 (1997), defending 'Megan's Law' requirement of registration
of sex offenders; Hutchins v. District of Columbia, 942 F.Supp. 665 (1996), defending
Juvenile Curfew Act; Herndon v. Chapel Hill-Carrboro City Board ofEducation, 89 F.3d 174
(1996), defending compulsory public service program for high school students;
Roulette v. City of Seattle, 78 F.3d 1425 (1996), supporting city laws against obstruction
of sidewalk; Montana v. Egelhoff 518 U.S. 37, 116 S.Ct. 2013 (1996), arguing against
defendant's right to introduce evidence of his voluntary intoxication to establish
diminished responsibility; Nunez v. City of San Diego, 963 F.Supp. 912 (1995),
supporting curfew for under-eighteens;Johnson v. City ofDallas, 61 F.3d 442 (5th Cir.
1995), supporting city laws against sleeping in public; Loperv. NYPD, 999 F.2d 699
(1993), supporting police action against panhandlers; and Michigan Department of State
Police v. Sitz, 496 U.S. 444, 110 S.Ct. 2481 (1990), defending highway sobriety
checkpoints.
3 Joyce v. City of San Francisco, 846 F. Supp. 843 (N.D. Calif. 1994).
4 There is an excellent account of the San Francisco Matrix Program and the public
response to it in N. Wright, 'Not in Anyone's Backyard: Ending the "Contest of
Nonresponsibility" and Implementing Longterm Solutions to Homelessness' (1995)
2 GeoJ. on Fighting Poverty 163 at 180-1.
5 J. Waldron, 'Homelessness and the Issue of Freedom' (1991) 39 U.C.L.A. Law Rev. 295
[hereinafter 'Homelessness']; reprinted in J. Waldron, Liberal Rights: Collected Papers
1981-1991 (Cambridge: Cambridge University Press, 1993) [hereinafter LiberalRights].
6 See especially Roulette v. City of Seattle, supra note 2, supporting city laws against
obstruction of sidewalk, and Johnsonv. City ofDallas, supra note 2, supporting city laws
against sleeping in public.

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
HOMELESSNESS AND COMMUNITY 373
Matrix program was an attempt, they said, to recover the public spaces of
the city for the community. So long as homeless people remain en-
camped in the city's streets, parks, and public squares, those places will
be cluttered with tents, dirty sleeping bags, cardboard shelters, and
stolen shopping carts and contaminated with urine, faeces, and drug
paraphernalia. Such conditions, argued the AARR, make it very difficult
for ordinary citizens, either individually or in families, to use those spaces
in the way that they were intended to be used. Panhandling, drinking,
and various forms of disturbed behaviour exacerbate the problem, mak-
ing the public urban environment not only unpleasant but hostile and
potentially dangerous. The result is that public places that used to be
available to the whole community are now 'becoming the preserve of
those on the margins of society.'7
The AARR argued that a community has a right to control behaviour in
its public spaces, and to outlaw activities such as drinking, panhandling,
sleeping on benches, washing in fountains, urinating and defecating in
public, and so on. The point of such restrictions, they said, is not to
oppress the homeless or to diminish their liberty, but to reduce annoy-
ance, to provide a fair basis 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
AARR argued that communities benefit from public spaces being kept
sufficiently attractive to act as public meeting places and as places where
people voluntarily spend their time.8 The brief spoke eloquently of a
time when citizens from all walks of life could spend their leisure 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 city to persevere in its efforts to restore this communi-
tarian mode of the use of its public spaces.
It seemed to me important to say, in response to these contentions,
that although it is certainly true (as the AARR put it) that 'governments
have the right to regulate certain types of conduct in public places, to
ensure that parks and sidewalks remain accessible and welcome to all,"'
and although a city must accept responsibility for maintaining the quality
of its public places and may not treat what happens there as a matter of
indifference, still the regulation of public space is a different matter in a
community some of whose members have no private space to retreat to
than in a community all of whose members have access to private spaces
- homes - as well as public spaces, where they can live their lives and take
7 Amicus brief ofAmerican Alliance for Rights and Responsibilities inJoyce v. City of San
Francisco, supra note 3 at 2 (on file with author.)
8 See Teir, Restoring Order, supra note 1 at 256.
9 AARR brief in Joyce, supra note 2 at 2.
10 Ibid. at 1.

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
care of their bodily needs. In the latter society, where everyone also has a
private place to go to, it is perhaps reasonable to say that the activities
performed in public might be the complement of activities appropriately
performed in private - a different and complementary set of activities.
But in the former society, where some individuals have no choice but to
live all their lives in public, that same complementarity cannot prevail. I
intend to elaborate that argument - about the implicit premise of
complementarity between public and private - which I think underlies
much communitarian writing (including, I fear, a certain amount of
communitarian writing by the left) about the use of public spaces, a little
later in the essay.
III Robert Ellickson
Before doing that, I would like to refer to a second occasion that has
stimulated my thoughts on these matters. This is a more academic
occasion: the publication in 1996 in the Yale LawJournal of a long article
by Yale law professor Robert Ellickson, entitled 'Controlling Chronic
Misconduct in City Spaces: Of Panhandlers, Skid Rows, and Public Space
Zoning.'" Professor Ellickson raised many of the same issues as the AARR
brief in Joyce;, but in a long law review piece he was able to address the
issues at greater length, to display the sinews of the argument, and to
offer readers a deeper opportunity for reflection on the premises of his
account. (I should add that Ellickson is interested not only in the
problem of regulating public spaces but in the more abstract issues that it
raises, for example, issues concerning the emergence and effectiveness of
social, as opposed to strictly legal, norms.)'2
Let me string together few quotations to give you a sense of the overall
flavour of Ellickson's article:
In large cities in the United States, government owns as much as 45% of the de-
veloped land area and allocate most of these public lands for use as streets and
highways.... To socialize its members, any society, and especially one as diverse as
the United States, requires venues where people of all backgrounds can rub
elbows.... A liberal society that aspires to ensure equality of opportunity and uni-
versal political participation must presumptively entitle every individual, even the
humblest, to enter all transportation corridors and open-access public spaces.
11 (1996) 105 Yale L.J. 1165 [hereinafter 'Controlling Chronic Misconduct'].

12 See also R.C. Ellickson, Order Without Law: How Neighbors Settle Disputes (Cambridge:
Harvard University Press, 1991) [hereinafter Order Without Law].I discuss this aspect of
Ellickson's work, in its bearing on the issue of street misconduct, infra in section XI.

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
A space that all can enter, however, is a space that each is tempted to abuse.
Societies therefore impose rules-of-the-road for public spaces.... Rules of proper
street behavior are not an impediment to freedom, but a foundation of it....
[T] o be truly public a space must be orderly enough to invite the entry of a large
majority of those who come to it. Just as disruptive forces at a town meeting may
lower citizen attendance, chronic panhandlers, bench squatters and other
disorderly people may deter some citizens from gathering in the agora."
The media are quick to report the gravest problems of the streets, such as armed
robberies, drug trafficking, and drive by shootings. This Article [Ellickson says]
focuses on problems that by comparison seem trivial: chronic street nuisances.
Chronic street nuisances occur when a person regularly behaves in a public pace
in a way that annoys - but no more than annoys - most other users, and persists
in doing so over a protracted period."4
And Ellickson offers this definition of the phenomenon he takes as his

target in the article:
A person perpetrates a chronic street nuisance by persistently acting in a public

space in a manner that violates prevailing community standards of behavior to
the significant cumulative annoyance of persons of ordinary sensibility who use
the same spaces.15
Unless chronic street nuisances of this sort are controlled, says

Ellickson, unless city dwellers can enjoy 'a basic minimum of decorum in
downtown public spaces, they will increasingly flee from those locations
to cyberspace, suburban malls, and private walled communities.''"6 The
rhetoric is more or less exactly that of the AARR, cited by the court in
Roulette v. City of Seattle. 'As amicus American Alliance for Rights and
Responsibilities explains on the city's behalf, "[a] downtown area
becomes dangerous to pedestrian safety and economic vitality when
individuals block the public sidewalks, thereby causing a steady cycle of
decline as residents and tourists go elsewhere to meet, shop and dine."'17
The reactions to Ellickson's article that I have seen's respond mostly
to a particular policy 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, 'Ellickson on "Chronic Misconduct" in Urba
Spaces: Of Panhandlers, Bench Squatters, and Day Laborers (1997) 32 Harv.C.R.-C.L.Law
Rev. 1. I am grateful to Professor Munzer for a most helpful conversation on these
issues. See also M.D. Rosen, 'Our Nonuniform Constitution: Geographical Variations
of Constitutional Requirements in the Aid of Community' (1999) 77 Tex.L.Rev. 1129
at 1135n; A.J. Gold, 'The Trinity Initiative in Economic Perspective: Place or People

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
system of zoning in public places, permitting panhandling and other

offensive behaviour at any time in certain zones that he calls 'Red Zones,'
permitting them on an intermittent but not chronic basis in other areas
that he calls 'Yellow Zones,' and forbidding them altogether in those
public places that he refers to as 'Green Zones.'9 I shall not say much
about that proposal, partly because it is not clear that Ellickson is offer-
ing it as anything other than a thought-experiment,20 but mainly because
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 particular zoning proposal.
IV Costs and benefits
What Ellickson's article and the AARR argument have in common is

concern about the offence and annoyance to ordinary citizens caused by
the presence and activities of homeless people in public places. (Profes-
sor Ellickson himself is not known as a communitarian theorist; though
there are clear communitarian themes in his work,21 his ideological
framework is more law-and-economics.) I want to begin my discussion by
Prosperity?' (1998) 30 Conn.L.Rev. 1317 at 1330n; R. Austin, '"NotJust for the Fun of
It!": Governmental Restraints 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 Welfare Reform' (1996) 6 U.C.L.A. Women's L.J. 427 at439n; G.E. Frug,
'City Services' (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
insight of Ellickson's zoning proposal takes its inspiration from the diversity of social
norms traditionally enforced in different parts of American cities: Because demands
on public spaces are highly diverse, city dwellers have historically tended to
differentiate their rules of conduct for specific sidewalks, parks, and plazas. Some
neighborhoods, like traditional Skid Rows, have been set aside as safe harbors for
disorderly people. Other sites, like tot-lots, have been allocated as refuges for persons
of delicate sensibility. A constitutional doctrine that compels a monolithic law of
public spaces is as silly as one that would compel a monolithic speed limit for all
streets. (Ibid. at 1247.)
20 Ellickson introduces his zoning proposal by saying: 'As a mental experiment, imagine
that it would be desirable for a city to have three codes, of varying stringency,
governing street behavior.' Ibid. at 1220 (my emphasis). This suggests that it is not to
be taken seriously as an actual policy recommendation. Elsewhere, however, Ellickson
refers to the idea that 'a city's codes of conduct should be allowed to vary spatially -
from street to street, from park to park, from sidewalk to sidewalk' as a 'central
normative thesis' of his article. Ibid. at 1171-2.
21 See ibid. at 1172, note 26. See also 'New Institutions For Old Neighborhoods' (1998)
48 Duke L.J. 75 at 107 ff, defending neighbourhood-level community organization
against liberal objections; and 'Property in Land' (1993) 102 Yale L.J. 1315 at 1344 ff,
on communitarian arguments about property.

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
considering the way in which he categorizes what he calls 'annoyance' in

an economic account.
Early in the article, Ellickson attempts what appears to be

mist's cost-benefit analysis of chronic street misconduct. Bo
dling and bench-squatting are said to confer benefits on var
As Ellickson puts it, in the peculiar language used in analyses
Panhandlers and at least some of their donors benefit from b

magnitude of these benefits depends on the opportunity costs inc
panhandling were to cease and both panhandlers and donors had t
their next best substitutes.22
For panhandlers, the benefit is the difference between be

(say) collecting cans and bottles; for those donors who 'take
pleasure in satisfying a request for a handout,' the extent of
turns on 'the quality of the almsgivers' other alternative
charitable.'23 Similarly, bench-squatting has economic benefit
[T]he magnitude depends on the quality of the bench squatter

alternatives. These might include: squatting in another public
suited to long-term stays; cycling among a number of public plac
ing more daylight hours indoors (perhaps in a board-and-care facili
center, a rented apartment, or a relative's home); and voluntarily in
tutionalization. Because the first alternatives listed are close substitu
fits of an entitlement to bench squat in a particular location ... ar
small.24
Those are the benefits. What about the harms? The harms, Ellickson
says, consist in 'minor annoyance' multiplied over many individuals and
sustained over a period of time. 'When being pan-handled,' he says, 'a
pedestrian of ordinary sensibility may feel some combination of: aggrava-
tion [sic] that his privacy has been disturbed, resentment that the pan-
handler's plea has a high probability of being fraudulent, and fear.'25
The fear element may not be present among those pedestrians who
encounter chronic panhandlers who are familiar to them; however,
[i]n other respects, ... the encounter may be more annoying than an encounter
with an unfamiliar panhandler, A pedestrian who sees a regular panhandler is
likely to become increasingly irked that the supplicant has not sought aid from
charities and welfare agencies better able than pedestrians to appraise desert.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 ana

25 Ibid. at 1181.
26 Ibid. at 1182.

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
378 UNIVERSITY OF TORONTO IAWJOURNAL
And such a panhandler may also annoy because his activity signals to
the pedestrian the breakdown of a social norm - such as the work ethic -
that the pedestrian cherishes.27 As for the annoyance caused by bench-
squatting, Ellickson says the following:
The most flagrant examples involve offense to a number of the senses. A man
who sits in a well-trafficked space amid shopping carts full of junk, who stinks
with body odor, and who urinates publicly into plastic jugs, is likely to trigger fre-
quent complaints to the police. A woman who sleeps on a busy sidewalk, who
smells of feces, and who shouts obscenities certainly engages in offensive
behavior....28
Now, these may all seem trivial harms - annoyances rather than
injuries. But Ellickson is right to emphasize that a set of harms should
not be ignored in policy analysis merely because each one considered by
itself is very small. As the moral philosopher Derek Parfit has empha-
sized, and as those who study collective action problems know,
consequentialist analysis can go seriously wrong by ignoring tiny harms.29
An individual automobile driver may release only small amounts of toxic
wastes into the environment when he drives to work. But the effects of
hundreds of thousands of drivers using that freeway on a given day may
add up to deadly pollution. Moreover, it may be impossible to solve such
a pollution problem except by subjecting all motorists to certain regula-
tions (e.g., requiring smog checks) and punishing them one by one -
despite the almost imperceptible harm that each causes individually - if
they do not comply with the regulations. So it is perfectly fair for Ellick-
son's analysis to take account of the fact that because a piece of street
misconduct (relatively trivial in itself) 'occurs in a public place, it may
affect hundreds or thousands of people per hour,' and for him to insist,
too, that 'as hours blend into days and weeks, the total annoyance
accumulates.'3o No doubt there are further questions to be asked about
how much moral weight we should give to the aggregation of tiny harms
in a social calculus. Do we ever want to say, for example, that a very great
harm (say, torture) inflicted on a single individual is the equivalent of
the sum of tiny harms (analogous to those Ellickson is studying) suffered
by N individuals (where N is a very large 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 outweighed by (say) thousands
of ambient annoyances suffered 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 Controlling Chronic Misconduct, supra note 12 at 1177.
31 See Parfit, supra note 31 at 75-82.

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
answer to the latter question is supposed to be 'Yes,' then what's the prin-
ciple underlying the difference? These are all interesting questions, and
they go to the heart of some central difficulties in utilitarian analysis.32
But the problem in Ellickson's analysis that I would like to explore arises
at an earlier stage.
v Are distress and annoyance harms?
The question I want to ask is whether the discomfort caused to ordinary

pedestrians by the presence and activities of homeless people should
even be considered a harm at all.
On the face of it, the question seems preposterous. Of course discom

fort and annoyance are harms. They may not be very serious harms
except to the extent that they accumulate - but surely they are to b
counted on the debit side of any plausible utilitarian calculus.
I am not so sure. Think of it along these lines. Imagine that in
country where homelessness has only recently become a problem
citizen previously unaware of the extent of the poverty in his societ
comes across a person living on the streets in filth and squalor. H
likely to be distressed by the spectacle, but his distress may have th
following flavour: 'This is awful. I am glad I have found out about th
and he may be moved by his horror to try and do something about it.3
this distress a harm to the citizen, something that in a utilitarian calc
should count pro tanto against his finding out about poverty and in fa
of his being sheltered from his knowledge? I think we should say n
not even that it is a slight harm out-balanced perhaps by greater bene
associated with his knowledge. On the contrary, there is a clear sens
which distress occasioned by the spectacle of another's suffering
good rather than an evil (just as pleasure occasioned by another
suffering is an evil, not a good). If there is first-order suffering of this s
around, then it is better that it be seen and that people be distressed b
than that it remain invisible to all but the immediate sufferers. A world
that differed from our own only in that the spectators of such suffering
were not moved to any sort of distress by the spectacle of poverty would
32 See also the discussion of trade-offs inJ. Waldron, 'Rights in 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 this pitiless 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 little care of this!

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
380 UNIVERSITY OF TORONTO IAWJOURNAL
be pro tanto a worse world. I shall call this the 'Appropriate Distress
Argument.'
Or think of the situation along these lines. A pedestrian comes across
a chronic bench-squatter, and he feels distress of the following kind: he
thinks to himself, furiously and agitatedly, 'It is outrageous that people like
this should sit idly around, instead of fulfilling their moral obligation to
get ajob and contribute to the social product.' But if someone asked why
the bench-squatter has a moral obligation to get ajob, it would surely be
inappropriate to cite the pedestrian's agitation as a utilitarian reason -
that is, 'People ought to get jobs so as to avoid distressing those who
think that they ought to get jobs.' To argue in that way would be like
arguing in a circle; one would be trying to defend a moral proposition by
citing forms of distress whose occurrence presupposed that the moral
proposition was true. I shall call this the 'External Preference Argument.'
Both arguments seem applicable to Ellickson's analysis - I mean
applicable as critiques. In both regards, his rather ingenuous account of
the 'cost' or the 'harm' caused by chronic misconduct in public places
ignores a lot of quite interesting work that has been done in recent
moral philosophy about the way in which preferences and putative harms
ought to be counted in a social calculus.
VI The Appropriate Distress Argument
There is a well-known controversy about the application of the 'Harm

Principle' propounded inJohn Stuart Mill's book On Liberty that goes as
follows. Suppose someone is distressed by the conduct of another -
suppose for example that a person feels offence and revulsion whenever
he sees two men kissing. Is distress of this kind - which is certainly painful
to experience - to be counted as harm for the purposes of a principle
which holds that 'the only purpose for which power can rightfully be
exercised over any member of a civilized community, against his will, is to
prevent harm to others'?34
The most plausible answer to that question - which I defended some
time ago in a little piece 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 intrinsically good for homophobes to suffer, but because it is
important for people to be confronted with ideas and ways of life that
challenge their own comfortable preconceptions. Ethical confrontation
- the confrontation of ideas and lifestyles - is a positive good for Mill. But
of course it is not a painless business. It hurts to be confronted in debate
34 J.S. Mill, On Liberty, ed. C.V. Shields (Indianapolis: Bobbs Merrill, 1956) at 13.
35 (1987) 35 Pol.Studies 410, reprinted in Liberal Rights, supra note 5 at 115.

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
if one takes seriously the views one is propounding, and it is distressing to

be faced with examples of lifestyles that pose a challenge to the founda-
tions of one's own. People are naturally upset when they are involved in
this sort of confrontation. If no one is disturbed, hurt, or distressed in
this way, that is a sign that vigorous ethical confrontation is not taking
place, and, for Mill, that in turn is a sign that the intellectual life and
ethical and cultural progress of our civilization may be grinding to a halt.
This is why Mill is very suspicious of any proviso to the effect 'that the
free expression of all opinions should be permitted on condition that the
manner be temperate':
If the test be offence to those whose opinions are attacked, I think experience
testifies that this offence is given whenever the attack is telling and powerful, and
that every opponent who pushes them hard, and whom they find it difficult to
answer, appears to them, if he shows any strong feeling on the subject, an
intemperate opponent.3"
Intemperance and the offence it inevitably occasions are indispens-

able for confrontation and for the progress that only confrontation can
bring about. Think what would be entailed by an interpretation that did
regard this sort of offence as sufficient to cross the threshold required by
the Harm Principle. On that assumption, what ought to be taken as evi-
dence that freedom of thought and lifestyle was promoting progress
would be invoked instead as a prima facie reason for interfering with that
freedom. A sign of vitality would be cited as a necessary condition for the
legitimate suppression of that vitality. A symptom of progress would
count as ajustification for acting in a way that would bring progress to a
halt. Mill cannot have held such a view.
Against this, it will be said that Mill was a utilitarian. How can
utilitarian not count discomfort, revulsion, and distress as pain?37 How
can a utilitarian not count them - as Professor Ellickson counts them -
on the debit or cost side of the social calculus? To answer this, we must
bear in mind the sort of utilitarian Mill was. Mill's utilitarianism is not a
Benthamite calculus of pleasures and pains, or of satisfactions and
dissatisfactions, of all sorts. The value on which liberty (defended by the
Harm Principle) is based is certainly utility, on Mill's account; but, as he
insists in the Introduction to On Liberty, 'it must be utility in the largest
sense, grounded on the permanent interests of man as a progressive
being.'38 And I take it that this passage refers not merely to the nature of
36 Mill, supra note 36 at 64-5.

37 This objection is put forward by T. Honderich, 'On Liberty and Morality-Dependent
Harms' (1982) 30 Pol.Studies 507.
38 Mill, supra note 36 at 15.

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
Mill's utilitarian computations - for example, taking a long-run rather

than a short-run view - but also to the content and character of his
utilitarian values. If we accept the arguments about the importanc
progress in Chapter 2 of On Liberty, or the arguments about sponta
in Chapter 3, it is not open to us to say that the distress experien
when one's preconceptions are challenged goes against one's interes
a progressive being. A creature who defined his interests - even in p
in terms of being free from the shock of ethical debate or free fr
anxiety about the grounds and worth of his lifestyle would be like
satisfied 'fool' in Mill's Utilitarianism.39
Now, I am not saying that Ellickson's pedestrians actually benefit by
being confronted with the unusual lifestyles of bench squatters and so on
(though they probably do). I have invoked Mill's argument primarily
because it opens up for consideration an array of ways in which we might
reject Ellickson's rather simple-minded characterization of all his pedes-
trians' distresses as harms.
The particular argument I want to make is that if Ellickson's pedestri-

ans are regarded as progressive beings, in Mill's sense, then we should
not go around saying that they have an interest in not knowing or in not
perceiving the true state of affairs in their society - the condition in
which some of their fellow citizens are having to live - simply because
that knowledge or perception is distressing to them. If the situation of
some in society is distressing, then it is important that others be dis-
tressed by it; if the situation of some in society is discomforting, then it is
important that others be discomforted.40
(By the same token, the distress occasioned by finding out some
unpleasant fact of our history should not be counted as a harm in
sophisticated social calculus; it seems crazy to say, as the tenor of Ellick-
son's analysis would seem to imply, that people of decent sensibility are
harmed by their distressing knowledge 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,42 then it is
39 'Utilitarianism' in J.S. Mill, Dissertations and Discussions: Political, Philosophical and

Historical, vol. 3 (New York: Henry Holt & Co., 1882) 300 at 312.
40 Thus I agree with Munzer, supra note 20 at 33, when he writes: 'Seeing the distress of
those who panhandle or bench squat is not, or at least not merely, an annoyance. It
is rather, or also, an experience of getting information, which no rational person
should consider a social harm, about the lives of a significant fraction of the poor in
America.'
41 But compare R.K. Fullinwider, 'Patriotic History' in R.K. Fullinwider, ed., Pu

Education in a Multicultural Society: Policy, Theory Critique (Cambridge: Cambri
University Press, 1996) 203.
42 J.-J. Rousseau, 'Preface' inJ.-J. Rousseau, A Discourse on the Origin ofInequality, tra
Coleman (New York: Oxford University Press, 1994) 3.

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
important that such repugnance not be treated as itself an evil on the

order of the first-order suffering that evokes it. Or, at the very least, it is
important that it not be treated as an independent evil, that is, as an evi
that is independently relievable or preventable. The only appropriate
way to relieve or prevent the suffering occasioned in me by seeing
another being suffer (which we may call 'respondent suffering') is to
relieve or prevent the first-order suffering to which it is a response. If, on
the contrary, one were to try to relieve or prevent respondent suffering
as an independent matter, one might do one's best to ensure that first-
order suffering took place as far as possible out of sight, unperceived by
those who might be moved by it. (Or one would try, by education or
conditioning, to deaden the affective response in people to such first-
order distress as remained visible.) But an attempt so motivated to
render first-order suffering invisible, or to block the normal response to
it, would be a way of showing that one did not understand the function
of respondent suffering and empathetic distress 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 precisely such a
misunderstanding - a misunderstanding that is not, I think, mitigated by
the use of such jargon terms as 'compassion fatigue.'43
Having said that, in all fairness, I should add that sympathetic distress
at the sight of the homeless is not Ellickson's main concern in the article
so far as the putative costs of street misconduct are concerned. He also
mentions revulsion - at body odours and the stink of urine and faeces -
and annoyance.
So far as revulsion at odours is concerned, the situation is compli-
cated. It is, of course, appropriate to recoil in public from the smell of
urine and faeces. This sort of sensory distress has a proper motivating
effect, namely to ensure as far as possible that human waste is disposed of
in sanitary conditions. The argument for providing public lavatories is
partly dependent on the importance of avoiding distress of this kind.44
But so far as the homeless are concerned, that is only part of the argu-
43 E.g., Controlling Chronic Misconduct, supra note 12 at 1168, 1178, 1218; the same
language is also used by the AARR, e.g., in Teir, 'Maintaining Safety and Civility, 'supra
note 1 at 288. For a critique of the use of this phrase, 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 provision of public lavatory facilities in the United States
and its impact 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 Georgetown J. on
Poverty Law & Policy 145 at 154; and L. Sossin, 'The Criminalization and
Administration of the Homeless: Notes on the Possibilities and Limits of Bureaucratic
Engagement' (1996) 22 N.Y.U.Rev.L.& Soc.Change 623 at 653.

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
ment. As I emphasized in my earlier piece, the main argument for this

provision has to do with the immediate distress and indignity suffered by
those who have to live almost their whole lives in a public space in which
such facilities are not provided.45 I will return to this later.46 For now, the
point I want to emphasize is that since a conscious political decision has
been made not to provide such facilities - this is true in the United States
(for reasons, I must say, that entirely escape me);47 it is not true in most
other advanced societies - then it is surely a bit much for those who have
failed to offer even this modicum of social provision to the homeless to
complain about the consequent smell and to treat that as grounds for
offering the homeless further indignities and restrictions.
VII Cognitive dissonance
What about annoyance? I think much of the angry side of the discomfi-
ture that concerns Ellickson is a matter of cognitive dissonance, associ-
ated with the visible refutation of the claims that Americans are proud to
make about their society.48 They say that theirs is a just and prosperous
society, a society of equal opportunity; they say that capitalism and the
market economy work wonderfully. But they are confronted on the
streets of their cities with hundreds or thousands of persons who live on
the very margins of civilized existence, on the very margins of life itself -
who flaunt in their persons the abject and desperate poverty that disfig-
ures American society and the idea of the American dream. Since their
feeling good about themselves as Americans depends primarily on a
sense that these things cannot be, ordinary people tend to respond with
annoyance and anger to the sight that they very evidently are. Now, I am
not making the argument that the United States should live up to its
ideals - that is almost certainly a hopeless enterprise. But political and
social theories that count as harm - for purposes of social policy - the
negative feelings of anger and annoyance that people experience when
they are shown that their society is not living up to its ideals - theories
that count this as harm, to be avoided if possible by removing the
spectacle that occasions the dissonance - are treading on very dangerous
ground. Since the Enlightenment, it has been a principle of good social
45 Waldron, 'Homelessness,' supra note 5 at 320-1 (Liberal Rights at 334-5).

46 See infra section X, especially notes 89-91 and accompanying text.
47 See C. Habermas, 'City Hall Can't Answer Nature's Call' New York Times (11 February
2000) Bi; A. Hagedorn & W. Green, 'Advocacy Group-Sues NewYork, Citing Lack of
Public Restrooms' Wall Street Journal (1 November 1990) B8. See also Lucasv. Dinkins,
608 N.Y.S.2d 403 (App.Div. 1994), holding that homeless individuals' complaint about
New York City's failure to provide public toilets was notjusticiable.
48 For a similar phenomenon in Canada, see J. Stackhouse, 'My Life Without a Home'
The [Toronto] Globe and Mail (18 December 1999) 'Focus' 1, and the ensuing
discussion.

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
and political theory that the social order should be subject to a constraint
of transparency - which means, roughly, that social stability should not
depend on any comprehensive misunderstanding of social reality on the
part of citizens.49 In a well-ordered society, asJohn Rawls puts it, 'nothing
need be hidden.'"5 Now, of course, we all know that, in the United States
at least, social stability does largely depend on misunderstanding and false
ideological perception of actually existing social conditions. But it is rare
to find a theorist making a virtue of it, let alone designating the annoy-
ance at having the illusion dispelled as a cost or a harm, to be prevented,
if need be, by the police and the criminal law.
Of course, Ellickson would not describe it in this way. He resorts
instead to the simple utilitarian formulas of the law-and-economics
movement. The flaw in his article is that he has not thought through the
significance of lumping all forms of distress together as social harms.
Law-and-economics types often present themselves as hard-headed men
of the world, willing to take a much clearer view of what is really going on
than most of their sentimental philosophical opponents.5' But this self-
presentation is undermined if what law-and-economics yields is a social
theory that panders, in the name of utility or of the minimization of
social cost, to people's desires not to have their illusions about the society
they live in dispelled.
VIII Broken windows
This may be the appropriate place to say something about the relation
between the visibility of homelessness and a theory of community
policing sometimes referred to as the 'Broken Windows Theory.'52 This is
49 SeeJ. Rawls, A Theory ofJustice (Cambridge: Harvard University Press, 1971) 133 at 454.
See also the discussion inJ. Waldron, 'Theoretical Foundations of Liberalism' (1987)
37 Phil.Q. 127 at 134-5 and 146 ff., reprinted in Liberal Rights, supra note 5, 35 at 43-4
and 57 ff.
50 J. Rawls, Political Liberalism (NewYork: Columbia University Press, 1983) at 68. Rawls
adds in a note: ' ... in a free society that all correctly recognize as just there is no need
for the illusions and delusions of ideology for society to work properly and for citizens
to accept it willingly. In this sense a well-ordered society may lack ideological or false
consciousness.' Ibid. at 68-9, note 21.
51 It was, after all, the grandfather of rational choice theory, Thomas Hobbes, who
invented and defended the principle of transparency. See Hobbes, Leviathan, ed.
Richard Tuck (Cambridge: Cambridge University Press, 1991) ch. 30 at 231-2. See also
J. Waldron, 'Hobbes on Truth and Civil Doctrine' in A. Rorty, ed., Philosophers on
Education (London: Routledge, 1998), and 'Hobbes and the Principle of Publicity' in
B. Honig, ed., Essays in Honor of Richard Flathman (forthcoming, 2001).
52 For good overviews see 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

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
the view put forward by George Kelling andJames Q. Wilson in a famous

article published in 1982, in which they argued that visible signs on the
streets of 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 likely to attract further disorder, and also crime,
because they signify to anyone who is interested that social controls are
weak or attenuated at that locale."5 According to some proponents of this
theory, derelict human beings behaving in an offensive fashion are just
one more such sign of disorder:
[J]ust as unrepaired broken windows in buildings may signal that nobody cares
and lead to additional vandalism and damage, so untended disorderly behavior
may also communicate that nobody cares (or that nobody can or will do
anything about disorder) and thus lead to increasingly aggressive criminal and
dangerous predatory behavior."54
On this account, 'the ill-smelling drunk [or] the unchecked panhan-

dler is, in effect, the first broken window.'55 Or, as Ellickson puts it, 'a
regular beggar is like an unrepaired broken window - a sign of the
absence of effective social-control mechanisms in that public space.'56
Now, I think we ought to be very watchful of theories of community
control that assimilate human beings even figuratively to things, like un-
collected garbage or shattered window panes. (If Kelling and Wilson had
titled their article 'Broken People,' would it perhaps have had a different
impact?) If we are to countenance policy talk about broken people on
the analogy of broken windows, then we need to ask two tough questions:
(1) Relative to what norms of order are bench-squatters or panhan-
dlers or smelly street people described as 'signs of disorder'? Are these
the norms of order for a complacent and self-righteous society, whose
more prosperous members are trying desperately to sustain various
delusions about the situation of the poor? Or are they norms of order for
a society whose members are attempting in good faith to live honestly
with a given mixture of great prosperity and great poverty? In some
societies whose members are familiar with great and endemic poverty -
India springs to mind as an example - people regard street begging as a
normal activity, 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' Atlantic Monthly (March 1982) 29.
54 G.L. Kelling, 'Acquiring a Taste for Order: The Community and Police' (1987) 33
Crim.& Delin. 90 at 92, quoted by AARR counsel Rob Teir in 'Maintaining Safety and
Civility,' supra note 1 at 290.
55 Kelling & Wilson, supra note 55 at 34.

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
plaint about such 'disorder' would be: 'Given the existence of poverty
(which you propose to do nothing about), what on earth do you expect?'
I belieye there is a dignity in that candour, certainly compared to the
shameful and self-righteous denial that we find on the streets of America
and the infantile preference for image over reality that dominates
discussion in the United States. So the first question is this: Are the
norms - relative to which human windows seem 'broken' - norms of
order for society in which it is envisaged that everyone has a home
to, that is, norms predicated on the assumption of complementarity
I mentioned in section III? Or are they norms of order based o
honest grasp of economic reality in an unequal society?
(2) The second question is this: Relative to whatever norms are ap
priate, what (according to the Kelling and Wilson approach) is to c
as fixing the window, when the 'broken window' is a human being
panhandler needs, or thinks he needs, money. Presumably, giving
money is not 'fixing the window' in terms of the Broken Win
Theory. But then what is? Suppose we are regularly confronted in a
lict neighbourhood with people who smell of urine and faeces. The
broken window, according to the Kelling and Wilson approach. Now
do we fix it? For some reason - in the United States at least - nobo
(certainly nobody who invokes this theory) seems to think that the
priate answer is: 'Provide public lavatories and public shower facili
Instead, fixing the window is taken to mean rousting the smelly in
ual and making him move out of the public park or city square. (I
though the smartest way to fix an actual broken window were to k
down the whole building, or move it to just outside the edge of tow
There is much more one could say about the Broken Windo
Theory. One could question, for example, the theory's determin
assumption of there being a constant potential for escalating, pred
crime, roaming the cities, looking for disorderly and neglected sit
settle on. One could talk about the way in which Ellickson, Teir
others have converted (or, more accurately, hijacked) a theory
policing priorities57 and turned it into a theory of legislation - th
into a theory about what sorts of things (e.g., panhandling, sleepin
public) should be made into offences. And, in that regard, one
dwell for a moment or two on the ethics of criminalizing acti
performed by one group of people in order to offset the attraction
their derelict condition may offer to other offenders."8
57 That is, a theory about the importance of the police focusing on minor as w
major crime: see, e.g., W.J. Bratton, 'The New York City Police Department'
Enforcement of Quality-of-Life Crimes' (1995) 42J. Law & Pol'y 447.
58 Cf. Foscarinis et al, supra note 46 at 153: 'This punitive approach raises s
concerns about fundamental fairness.... [P]unishing one group of people to p

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
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 with the
devil. For decades we conjectured that poverty for some would lead
eventually to a deterioration in the quality of life for everyone, even for
the rich and comfortable. And on the basis of that conjecture we sought
to mitigate the worst effects of inequality. We did so in our own interest,
as well as on the basis of more altruistic and social justice concerns. We
believed that if we didn't, inequality would eventually redound to the
detriment of us all. Since 1980, however, the United Kingdom first, then
the United States, and then other countries following their lead have
decided to test that conjecture and, if possible, refute it. (This is the
devil's bargain I refer to.) Now we are working on a different hypothesis:
maybe extreme poverty for a substantial section of society can be toler-
ated with impunity, without undermining (even in the long run) security
and quality of life for the most prosperous and the opportunities they
cherish for their children. Maybe aggressive policing strategies mean that
we can have all the glamour of a prosperous-lookingsociety without doing
very much - 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 hypothesis is true. Maybe the
bargain with the devil will work, with a huge payoff - a sort of
compassion-fatigue dividend - for those of us who are prospering.
Maybe. But it does seem a bit much - it seems unconscionable, in fact
- to complain about and stigmatize the poorest of the poor for refusing
to cooperate with this experiment in the cosmetics of injustice, for
refusing to play their part in this new scenario in which we tolerate and
benefit from inequality while protecting our illusions and sensibilities. It
does seem a bit much to characterize the poorest of the poor as 'broken
windows' relative to a self-image of righteous prosperity and order, when
it is not entirely clear - indeed, when it is exactly the point of our
experiment to see - whether we are entitled to project that image."5
future criminal activity by others runs afoul of the basic notions of equity underlying
our criminal justice system.'
59 In times past, inequality and great poverty have themselves been regarded as social
disorders, as 'broken windows' in the fabric of a well-ordered society; and it has been
thought that sweeping them from view, far from remedying the disorder, actually
compounds it. Here I agree with Munzer, supra note 21 at 33, when he writes, 'The
existence of what Ellickson calls "street disorder" has not merely a corrosive effect on
the public good of harmonious city life but also is an indicator of social injustice. It
serves as a constant reminder that something is deeply wrong with a society that has
vast numbers of bench squatters and chronic panhandlers.'

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
IX The External Preference Argument
I mentioned two lines of argument against Ellickson's characterization of

offence and annoyance as social costs. The first was the 'Appropriate
Distress' argument, which I set out in section VII. The second line of
argument I called the 'External Preference Argument,' and it pertains to
Ellickson's claim that '[a] pedestrian who sees a regular panhandler is
likely to become increasingly irked that the supplicant has not sought aid
from charities and welfare agencies better able than pedestrians to
appraise desert'" and to his view that this 'irk' or this being irked consti-
tutes a harm to the 'irkee.'
The External Preference Argument was made originally by Ronald
Dworkin to criticize versions of the utilitarian calculus that treated
people's preferences as to what others should do or have (which D
called 'external preferences') on a par with people's preferen
themselves ('personal preferences')."6 In the debate about seg
education, for example, Dworkin believed that it was approp
consider both the costs and the benefits of integration, but he th
was not appropriate to count, as a cost of integration, the frustra
the desire of some racists that blacks count for less or that they b
equal educational opportunities. Counting such preferences
undermine whatever moral appeal utilitarian calculations have as
basis for the evaluation of costs and benefits. One ground for the
appeal of the utilitarian calculus is the utilitarian's commitm
equality, at least at the level of inputs: each person is to count f
nobody for more than one.
If a utilitarian argument counts external preferences along with
preferences, then the egalitarian character of that argument
rupted, because the chance that anyone's preferences have to
will then depend, not on the demands that the personal preferen
others make on scarce resources, but on the respect or affection
have for him or his way of life. If external preferences tip the b
then the fact that a policy makes the community better off in a utili
sense would not provide ajustification compatible with the right o
it disadvantages to be treated as equals."62
I will not go into what Dworkin has made of this point or the
role it plays in his theory of rights-as-trumps-over-utility."3 I do
60 Controlling Chronic Misconduct, supra note 13 at 1182.

61 R. Dworkin, Taking Rights Seriously, rev. ed. (London: Duckworth, 1977) at 2
275-6.
62 Ibid. at 235.
63 But see ibid. at 274-8 and 356-9. The connection is set out in J. Waldron, 'Pildes
Dworkin's Theory of Rights' (2000) 29J.Legal Stud. 301.

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
emphasize, however, that it applies not only to vicious external prefer-

ences like the racist onesjust mentioned but to virtuous, even admirable,
external preferences as well. In a notable critique of Dworkin's argu-
ment, H.L.A. Hart argued that it was in fact perfectly appropriate for
external preferences to be counted in the determination of social
policy."4 Hart argued that the importance of not disqualifying external
preferences is shown by the example of homosexual law reform in
England in the 1960s, where, he says, it is perfectly possible that 'it was the
disinterested external preferences of liberal heterosexuals that homo-
sexuals should have this freedom that tipped the balance against the
external preferences of other heterosexuals who would deny this free-
dom.'"6 How could anyone possibly complain about counting external pre-
ferences in that case? And if we count them in that case, why not count
them in any case, including cases of the sort that Ellickson describes?
Hart's critique has convinced many commentators.66 But I think
Dworkin's response to Hart is exactly right, and it is adaptable to the case
we are considering. Dworkin concedes that a simple-minded form of
utilitarianism would count external preferences: it 'would count the
attractive political convictions of the liberals of the nineteen-sixties
simply as data, to be balanced against the less attractive convictions of
others, to see which carried the day in the contest of number and
intensity.'"67 But this, he says, is surely not how the liberals of the 1960s
intended their preferences to be taken:
They of course expressed their own political preferences in their votes and
arguments, but they did not appeal to the popularity of these preferences as
providing an argument in itself for what they wanted, as the unrestricted
utilitarian argument I oppose would have encouraged them to do."8
The arguments made by the liberal supporters of gay rights were

presumably arguments of principle - arguments (e.g., about liberty and
privacy) intended to stand on their own grounds, and not on the mere
64 H.L.A. Hart, 'Between Utility and Rights' in A. Ryan, ed., The Idea ofFreedom: Essays in
Honor of Isaiah Berlin (Oxford: Oxford University Press, 1979) 77 at 86 ff.
65 Ibid. at 92-3.
66 See, e.g., J.H. Ely, 'Professor Dworkin's External/personal Preference Distinctio

(1983) Duke L.J. 959; T. Morawetz, 'Persons Without History: Liberal Theory an
Human Experience' (1986) 66 B.U.L.Rev. 1013 at 1017; and P.E. Johnson, 'Do Yo
Sincerely Want to Be Radical?' (1984) 36 Stanford L.Rev. 247 at 276. However, for
more careful view of Hart's critique, see J. Raz, 'Liberalism, Skepticism, an
Democracy' (1989) 74 Iowa L.R. 761 at 772-4.
67 R. Dworkin, 'Rights as Trumps' inJ. Waldron, ed., Theories of Rights (Oxford: Oxfor
University Press, 1984) 153 at 162.
68 Ibid.

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
fact of their being held and supported. Certainly Dworkin is not arguing
that people should refrain from expressing their external preferences or
their principled convictions; he is not urging a politics of self-interest."6
His argument is, rather, that the utilitarian calculus has specific work to
do in politics, and that when it does its work it should work in the
domain of personal preferences to establish who would be harmed or
who would be benefited in terms of what they want for themselves from
some political proposal. That, then, may be used as a way of defending or
criticizing political proposals, and it may be appealed to as a basis for
voting one way or the other.7" But it can do this work only if it is con-
ceived as a distinct form of political justification, not if it operates as a
function over moral convictions that people already have and that they
justify (if they do) on independent grounds.
The point, then, is this. In cases where people feel vehemently about
some matter of principle - whether they feel great because the principle
69 See Dworkin, Taking Rights Seriously, supra note 63 at 358: 'Nothing could be further
from what I suppose than the idea that people should act only in their own interests
and never in the interests of children, lovers, friends or humanity, or that their votes
should not represent their ideals of justice or other political ideals as well as their
selfish interests.'
70 The connection between votes and external preferences is subtle and important.
People inevitably - and properly - vote on the basis of their external 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 constraints about how far preference
utilitarianism provides ajustification for their decisions; that is, how far the fact that
a majority prefers a particular state of affairs (as distinct from the justice of what the
majority wants) counts as an argument for a political decision to promote it. (Ibid.)
This passage needs to be read carefully. In the context of a plebiscite, the fact that
a majority support a proposal makes it legitimate; and in the context of an election,
the fact that a majority shows support for a policy provides an electoral mandate for
it. But in neither of these cases does it necessarily provide an argument for it (e.g., an
argument that might convince a citizen to vote one way rather than the other).
Whether it provides an argument depends on whether something like utilitarianism
is an acceptable theory of social policy in a given area, and it is in relation to that
question that Dworkin's exclusion of external preferences is important.
To put it another way, voting doesn't purport to justify a political position (although
it may, in some circumstances, be cited as evidence in a utilitarian justification).
Rather, it is a way of choosing political positions for society in circumstances where
people disagree about what is justified. But utilitarianism does purport to be a
justificatory theory, and in that context there is something very fishy about citing as
part of the justification of a principle a preference or feeling whose felt character
presupposes that the principle is justified. For further discussion of this distinction
between first-level arguments and second-level legitimacy, see J. Waldron, 'The
Circumstances of Integrity' (1997) 3 Leg.Theory 1 at 9-12, andJ. Waldron, Law and
Disagreement (Oxford: Clarendon Press, 1999) at 195-8.

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
has been vindicated or whether they feel offended because it is violated

- it almost always misrepresents 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 view of the principle's justifica-
tion, these sorts of pains and pleasures are mere epiphenomena. Which
is not to say that the principle may not have a utilitarian justification; but
it is not ajustification of this sort.
I made this deviation into the controversy about external preferences
because, again, I wanted to indicate how poorly thought through Profes-
sor Ellickson's conception of the harms occasioned by panhandling and
bench-squatting actually is. We have already seen that the annoyance
aspect of being confronted by someone who casts doubt on one's view of
what society is really like should not be counted as a harm. If anything it
is a benefit, although it is probably more accurate to say that the simple-
minded cost-benefit dichotomy is quite inadequate to deal with all this.
And now we have seen that there is reason, too, for doubting whether
the normative, condemnatory side of the annoyance suffered by Ellick-
son's pedestrian should be regarded as harm for the purposes of develop-
ing anything like a utilitarian justification of restrictions on the homeless.
The pedestrian is irked that the bench-squatter is idle. He does not
think such idleness should be permitted, so he campaigns for a law
against bench-squatting, together perhaps with a 'workfare' requirement,
which would effectively put an end to the practice. Now, there may be
good reasons in favour of that legal change. If the pedestrian is aware of
them, then they are presumably the reasons that weigh with him, ex-
plaining why the bench-squatting annoys him in the first place. But the
self-aware pedestrian will not think that his annoyance justifies the pro-
posal to change the law; instead, he will think that the reasons that justify
the annoyance also justify the proposal.
Suppose now that the case in favour of the workfare-and-anti-bench-
squatting law is finely balanced. On the one side is the array of reasons -
R1, R2, R3 - that convinced the pedestrian; on the other side are various
reasons of a humanitarian and administrative character - call them Rs, Re,
and R7. When the issue is finely balanced like this, should we introduce
the fact that the proposed law would relieve the pedestrian's annoyance
as an additional reason R4, operating perhaps as a tie-breaker? Surely not.
The pedestrian's annoyance is not an additional factor in favour of the
law over and above R1, R2,, and R3. Rather, the pedestrian's annoyance is
simply the effect that R1, R2, and R, have in the mind of someone who
believes that they are not being properly attended to.
This is not to say that the condemnatory aspect of the anger that
Ellickson's pedestrians feel is unimportant or not worth considering. It is
important. But what is worth considering about it is the question of its

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
justification. If Ellickson's pedestrian is irked by the fact that homeless

people are violating rules of conduct for public spaces, then we need to
ask what these rules are and how are they justified; we should not be in
the business of citing the fact that the pedestrians are irked as a reason in
support of their views about these rules.
The External Preference argument may also work in another way, in
the contexts Ellickson considers. Many of those who are annoyed by the
demands of street people were brought up to believe that generosity is a
virtue, and that there is something wrong about simply turning one's
back on a cry for alms.71 Now, though, when the demands for help seem
overwhelming, the response is often to oppose to the felt duty of almsgiv-
ing a moral judgement to the effect that the poor must take responsibil-
ity for their own predicament. And so one walks past the beggars on the
street, repeating in one's mind furiously the modem mantra 'Own fault.
Own fault.' The opposition between these judgements may be experi-
enced subjectively as anxiety and agitation, as new economic wisdom
clashes with moral upbringing. Most of us feel bad after each of these
encounters, or each of these turnings-away. But it would be wrong to cite
the feeling as one of the costs of street misconduct. It is not an independ-
ent factor in the moral calculation; at best, it is one of the symptoms of
the moral calculation's being performed.
71 For the Christian view, see Matthew 25:31-46:
When the Son of man shall come in his glory, ... before him shall be gathered all
nations: and he shall separate them one from another, as a shepherd divideth his
sheep from the goats: And he shall set the sheep on his right hand, but the goats on
the left. Then shall the King say unto them on his right hand, Come, ye blessed of my
Father, inherit the kingdom prepared for you from the 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 visited me:
I was in prison, and ye came unto me. Then shall the righteous answer 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 answer and say
unto them, Verily I say unto you, Inasmuch as ye have done it unto one of the least of
these my brethren, ye have done it unto me. Then shall he say also unto them on the
left hand, Depart from me, ye cursed, into everlasting fire, prepared for the 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 they also 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 minister unto thee? Then shall he answer them, 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.

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
x Complementarity again
So I return to my little contretemps with the American Alliance for

Rights and Responsibilities. The premise of their position was that
'governments have the right to regulate certain types of conduct in
public places, to ensure that parks and sidewalks remain accessible and
welcome to all.'72 And as I said in section II, I agree with this. The ques-
tion is: How? On what basis is this regulation to be set up?
If there were no homeless persons in our community - that is, if
everyone living in our cities had access to private accommodation or
guaranteed shelter space for sleeping and for care of self, and so on -
then public spaces could be regulated on the following basis. Since
everyone would have access to a private home, activities deemed particu-
larly appropriate to the private realm - activities like sleeping, copulat-
ing, washing, urinating, and so forth - could be confined to that realm.
Public places could be put off-limits to such activities, and dedicated
instead to activities that complemented those that citizens performed in
their own homes. (This is the Complementarity Thesis I mentioned at
the beginning of this essay.)"7 Public places could be dedicated to things
like strolling, picnics, meeting people, walking dogs, children's play, 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 storing their
possessions there, and not to find evidence of human urination or
defecation. They could reasonably assume that everyone had a home to
go to for activities of that kind. Laws and regulations prohibiting sleeping
and storing possessions on sidewalks and public parks could be enforced
in the light of that expectation without fear of disparate impact.
That, in my view, captures the spirit of the AARR brief, which (as I said
earlier) evoked a time when citizens from all walks of life - ' [t]hose with
Armani suits, and those with nose rings; elderly people and gay couples;
residents and visitors; rich, middle, and struggling classes'74 - spent their
leisure hours in public places.75 It is an attractive idea, of public places
facilitating interaction among strangers, not just among friends,76 and as
such it is preferable to the usual nostalgia for Gemeinschaft that one finds
in the communitarian literature.77 As AARR counsel Rob Teir puts it,
72 AARR brief inJoyce, supra note 7 at 1.

73 Towards the end of section II, supra.
74 Teir, 'Restoring Order,' supra note 1 at 290.
75 AARR brief in Joyce, supra note 7 at 2.
76 See also the discussion in M. Ignatieff, The Needs of Strangers (London: Chatto &
Windus, 1984) at 139 ff.
77 See also J. Waldron, 'Particular Values and Critical Morality' 77 Cal.L.Rev. 561 at 581
ff., reprinted in Liberal Rights, supra note 5, 168 at 193 ff.

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
City parks and sidewalks were built to be community meeting places, where
people of different races, religions, ethnic groups, socio-economic levels, and
political views, could come together and share in the benefits of public spaces.
These venues are places of integration, assimilation, mixture of social classes,
and a counterweight to the increasing fragmentation of society.78
We are to imagine diverse citizens coming out into the parks and
boulevards where they can enjoy their leisure, 'interact with their fellow
citizens and leave behind their isolation and segregation'79 before
returning once again to the private realm.Certainly this is an attractive
image. Unfortunately, it is not appropriate for the regulation of public
places in a society where there are large numbers of homeless people. In
such a society, public spaces have to be regulated on a somewhat differ-
ent basis. They have to be regulated in light of the recognition that some
people have no private space - not even the temporary privacy that
public shelters or public toilets would afford - to come out of or to
return to. Fairness demands that public spaces be regulated in light of
the recognition that large numbers of people have no alternative but to
be and remain and live all their lives in public. For such persons, there is
an unavoidable failure of the complementarity between the use of
private space and the use of public space, and unless we are prepared to
embrace the most egregious unfairness in the way our community polices
itself in public, we are simply not in a position to use that complementar-
ity as a basis for regulation."
It is worth dwelling on the question of fairness. Professor Ellickson
recognizes it as a formidable issue,"s but the AARR's counsel Robert Teir
protests that
78 'Maintaining Safety and Civility,' supra note 1 at 289.

79 Ibid. at 337.
80 Whether this works as a constitutional argument is another matter: one has to contrast
the argument that worked in Pottinger v. City of Miami, 810 F.Supp. 1551 at 1571
(S.D.Fla. 1992) - holding that the 'practice of arresting homeless individuals for
performing inoffensive conduct in public when they have 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 infringes on the fundamental right to travel in violation of the equal
protection clause of the fourteenth amendment' (ibid. at 1583) - with the argument
that was rejected in Joyce v. City of San Francisco, 846 F.Supp. 843 (N.D.Calif., 1994).
But even assuming that a constitutional challenge fails, the fact that the legislation
is constitutional does not mean it is fair. This essay should be read as a consideration
of the issue of fairness without much regard to the political forum in which that issue
might be explored.
81 See 'Controlling Chronic Misconduct,' supra note 12 at 1247: 'The reconciliation of
individual rights and community values on the streets is a profoundly difficult
problem.'

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
[t] here is nothing unfair or mean-spirited about wanting to be free from harass-
ment and intimidation, wanting urban parks where children can play and adults
can enjoy the green, and the quiet, or wanting urban parks that are not filled with
litter, human waste, needles, bedrolls, drug users, and used condoms.... These
rules ... are set so that all people feel welcome in the public spaces.82
He is right. In itself, the aspiration he mentions is just and admirable.

Unfairness comes into the picture only when we consider the implemen-
tation of this goal against the background of homelessness. But again
Teir protests. He says that those who suffer from the deterioration of
public places are often the 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 favour of the rich; at worst, it is discrimination against
one class of poor people in favour of another.8" And anyway, he says, the
measures discussed in this article are not aimed at the homeless. Rather,
they are aimed at and address conduct, and only those who choose to
engage in the prohibited conduct fall within their reach.84
Once again, he may be right - at least as to the intention.85 But one is
entitled to- consider, also, evident and predictable disparities that
82 'Restoring Order,' 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 affluent who reap the
benefits of these measures. The rich, after all, can take care of themselves. They are
not, speaking generally, dependent upon public parks for recreation. They usually live
in secured communities, and shop in safe and comfortable places. The well-off can
also leave an area when it gets intolerable. Rather, it is the poor and middle-classes
who depend upon the safety and civility of public spaces. They have fewer options
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.
observers concerned primarilywith distributivejustice, extreme poverty might furnis
... a sufficient reason for siding with a disorderly street person in any policy cont
This is an ill-considered position. To favor the poorest may disadvantage the p
who are as unhappy with street disorder as the rest of the population. Beca
residents of poor urban neighborhoods tend to make especially heavy use of stre
and sidewalks for social interactions, they have an unusually large stake in prevent
misconduct there.
These are perfectly reasonable points. The issues about fairness discussed in the tex
concern, not rich/poor comparisons, but the enormous difference in burden cast b
public-space regulations on those who are homeless and those (whether rich or poor
who are not.
84 Teir, 'Restoring Order,' supra note 1 at 291.

85 For some doubts, see Waldron, 'Homelessness,' supra note5 at 313-4 (Liberal Rights
at 327-8), setting out four reasons for doubting the impartiality of the intentions of
those who lobby for prohibitions on sleeping in public.

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
different classes of people will experience in bringing their behaviour

within the norms he wants to enforce. If not as a constitutional matter,
then certainly as a matter of justice, those who have the power to regu-
late public places must pay special attention to the difference between
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 with an ordinance prohibiting, for example, sleeping
in public places is simply a matter of relocation. For someone who has no
home, however, and - let's say - no access to a shelter,86 compliance with
such an ordinance would mean that he must not sleep (for there is now no
place where his sleeping is permissible).87 This may not be what any
enforcer or communitarian lobbyist intends, but it is the easily foresee-
able result of a number of familiar intentional prohibitions.88 This
impact is so qualitatively different from the impact of the regulation on
the person who has a home to return to that it amounts almost to the
application of a quite different set 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 proportion to the minor inconve-
nience that would be suffered by a person who had somewhere else to
sleep. And, of course, eventually, the effective prohibition on sleep
becomes physiologically impossible to comply with.81' (Against all this,
86 More about this condition in a moment: see text accompanying note 93 infra.
87 Waldron, 'Homelessness,' supra note 6 at 315 (Liberal Rights at 328-9):
For a person who has no home, and has no expectation of being allowed into
something like a private office building or a restaurant, prohibitions on things like
sleeping that apply particularly to public places pose a special problem. For although
there is no general prohibition on acts of these types, still they are effectively ruled out
altogether for anyone who is homeless and who has no shelter to go to. The
prohibition is comprehensive in effect because of the cumulation, in the case of the
homeless, of a number of different bans, differently imposed. The rules of property
prohibit the homeless person from doing any of these acts in private, since there is no
private place that he has a right to be. And the rules governing public 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 private places and public places between
them exhaust all the places that there are, there is nowhere that these actions may be
performed by the homeless person. And since freedom to perform a concrete action
requires freedom to perform it at some place, it follows that the homeless person does
not have the freedom to perform them. If sleeping is prohibited in public places, then
sleeping is comprehensively prohibited to the homeless.
88 For the detailed argument here, see ibid. at 315-7 (Liberal Rights at 329-32).
89 Ibid. at 320 (Liberal Rights at 334-5): '[A]ny restriction on the performance of these
basic acts has the feature of being not only uncomfortable and degrading, but more
or less literally unbearable for the people concerned. People need sleep, for example,
not just in the sense that sleep is necessary for health, but also in the sense that they
will eventually fall asleep or drop from exhaustion if it is denied them. People simply
cannot bear a lack of sleep, and they will do themselves a great deal of damage trying
to bear it. The same, obviously, is true of bodily functions like urinating and

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
Ellickson's observation that '[w] hile no one's will is fully free, virtually all
of us have some capacity for self-control' does seem a little lame, to say
the least.) '0
I said a moment ago that we should assume for the sake of argument
that the homeless have no publicly provided shelter to go to. For many of
the homeless this is false,91 although it is true during daylight hours, and
importantly true then so far as publicly provided bathrooms are con-
cerned. I suggested this assumption - artificial as it is - in order to
emphasize the extent to which the fairness of public-place regulations
depends on other aspects of provision for the homeless. Robert Teir
believes that the two issues are separable. He cites a comment by colum-
nist George Will to the effect that '[t] he question of what society owes in
compassionate help to street people is, surely, severable from the ques-
defecating. These are things that people simply have to do; any attempt voluntarily to
refrain from doing them is at once painful, dangerous, and finally impossible.' This
- as I said in the 'Homelessness' article (ibid. at 320; Liberal Rights at 334) - is
something that every torturer knows: '[T]o break the human spirit, focus the mind of
the victim through petty restrictions pitilessly imposed on the banal necessities of
human life. We should be ashamed thatwe have allowed our laws of public and private
property to reduce a million or more citizens to something approaching this level of
degradation. Increasingly, in the way we organize common property, we have done all
we can to prevent people from taking care of these elementary needs themselves,
quietly, with dignity, as ordinary human beings.'
90 Ellickson, 'Controlling Chronic Misconduct,' supra note 6 at 1187. The full text of the
passage reads (footnotes omitted):
Many advocates sincerely believe that street people are so constrained by economic
and social circumstances that they lack real choices. Most (although not all) social-
welfare professionals hold the view that poor people always act under duress;
according to this view, society should not 'blame' poor people or, under an extreme
formulation, ask them to bear any responsibilities. While no one's will is fully free,
virtually all of us have some capacity for self-control. Legal and ethical systems
therefore properly subscribe to the proposition - or salutary myth - that an individual
is generally responsible for his behavior. This policy, at the margin, helps foster civic
rectitude. To treat the destitute as choiceless underestimates their capacities and, by
failing to regard them as ordinary people, risks denying them full humanity.
91 For example, New York City guarantees a bed in a shelter to any person who seeks it
(subject to some work and conduct requirements): see P.T. Kilborn, 'Gimme Shelter:
Same Song, New Tune' The New York Times (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
Government Contraction' (1989) 104 Pol.Sci.Quar. 607.) A report commissioned by
the United States Conference of Mayors between 1986 and 1995 in twenty-nine major
cities indicates that in 1995, for example, 24% of shelter requests went unmet during
that year and that emergency shelters in 79% of the surveyed cities turn away
homeless families because of a lack of resources. See L.D. Waxman, K. Peterson, & M.
McClure, U.S. Conference of Mayors, a Status Report on Hunger and Homelessness in
America's Cities: 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.

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
tion of what right the community has to protect a minimally civilized

ambience in public spaces.'"2 The analysis I have given shows why the
questions are not severable: the less the society provides in the way of
public assistance, the more unfair is its enforcement of norms for public
places that depend on a complementarity that simply doesn't apply to a
considerable number of citizens. Notice, though, that this does not mean
that all regulation of public places must be suspended until provision has
been made for the needs of the homeless. From what I have said here,
nothing follows about the enforcement in public places of those rules of
conduct that apply everywhere: murder and assault will remain illegal in
public as well as in private, and if the use and supply of narcotics contin-
ues to be banned, then they too will be banned in public places. The
arguments I have made primarily concern necessary activities - such as
sleeping and urinating - which would be perfectly legal if performed in
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 that activity' - if we have not at the same
time taken steps to ensure that everyone has access to a private place in
which to perform them. So long as we have not taken those steps, then
banning such activities in public is like enforcing a curfew against the
homeless when there is nowhere for them to go. It fails to take proper
account of the fact that, as things stand, public places are the only places
they are allowed to be.
If, as a result, the visibility of these properly private (but, for the
homeless, unavoidably public) activities is a deterrent to more traditional
uses of public space by citizens who do have a home to go, then we have
no choice but to say to the latter, 'In the present state of public policy, we
can longer guarantee the use of public space for this sort of respectable
activity alone. Any use you make of the public space will just 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
your juggling, or play pick-up football, or lay out your picnic, or make
your promenade.' Now, this is certainly a matter of regret, and the
consequence may well be, as Ellickson has suggested, an impoverishment
of the public dimension of culture and civil society, as those who have a
choice flee the downtown streets and parks and take refuge in cyber-
space, suburban malls, or gated communities, leaving public places to the
mercy of those who have no option about remaining there.93 But it is
important to see that this is not the sort of dilemma that we can solve by
simply adjusting the regulations. We can't have it all ways: unless we are
willing to take the next step and actually eradicate the homeless or put
92 G.F. Will, 'Beggars and Judicial Imperialism' The Washington Post (1 February 1990)
A21, cited in Teir, 'Maintaining Safety and Civility,' supra note 1 at 337.
93 Ellickson, 'Controlling Chronic Misconduct,' supra note 12 at 1172.

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
them out of sight in concentration camps,'94 then they and their lives are
just there - unavoidably - by virtue of the fact that distribution of housing
has left them nowhere else to go. And our use and regulation of public
places must be adjusted accordingly.
I suspect that an apprehension of this collapse of the complementarity
on which the traditional use of public space depended lies at the heart of
many of the attitudes that Robert Ellickson discerns in his pedestrians.
The rules that concern them - the rules they are distressed to see the
homeless flouting - are predicated on the traditional public/private
complementarity; and the distress they feel may reflect an angry frustra-
tion arising from their knowledge that the basis for the fair enforcement
of those rules no longer exists. If this is true, then there is an additional
mistake in Ellickson's citing such frustration as a utilitarian ground for
persisting with the rules in question, or enhancing or intensifying their
enforcement. For now we see that the very distress he regards as a harm
of the homeless people's activity is in fact frustration associated with an
apprehension of the unfairness of effectively prohibiting that activity,
indeed the moral impossibility of the whole traditional framework of
rules of that kind. Once again we see how badly one can go wrong by
simply categorizing every negative emotion that people feel as a cost for
the purpose of an economic calculus.
XI Social norms
Let me head for the finish on a more conciliatory note. The best work
that Ellickson has done concerns the emergence and role of social norms
- informal customs and practices addressing problems and conflict on
the ground, so to speak, to which formal state or sovereign law often
takes a distant second place."9 In his 1996 article on 'Chronic Misconduct
in Public Places,' Ellickson recognized that this social norms perspective
94 For apprehensions along these lines see, e.g., Bob Pool, 'Fanfare, Fear Surround New
Shelter' Los Angeles Times (16 April 1999) at Bl:
Development of the center was first proposed in 1994 by downtown business leaders
and supported by Mayor Richard Riordan. Its original concept called for a $4 million
urban campground serving as many as 800 homeless people on a fenced-in lot. As part
of that plan, outreach vans would circulate through downtown streets and social
services workers would invite transients to ride with them 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 Order Without Law, supra note 13, discussing informal 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 Stanford L.Rev. 623
and 'Law and Economics Discovers Social Norms' (1998) 27J.Legal Stud. 537.

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
ought to apply to city streets also. He invoked an oft-cited dictum from

Jane Jacobs's classic study of urban organization:
[T]he public peace - the sidewalk and street peace - of cities is not kept
primarily by the police, necessary as police are. It is kept primarily by an
intricate, almost unconscious, network of voluntary controls and standards
among the people themselves, and enforced by the people themselves.... No
amount of police can enforce civilization where the normal, casual enforcement
of it has broken down.96
He also expressed some apprehension about the obstacles that defeat

or undermine this pattern of 'normal, casual enforcement,' so far as the
prevention of chronic annoyance by the homeless is concerned. In
principle, it should be possible, he said, 'for a bystander to intervene to
prevent a street person from annoying another sidewalk user.'97 Even a
person who is 'reluctant to chastise a panhandler may be willing to frown
at an almsgiver.'98 But, he laments, these things seldom happen. His
explanation has to do mainly With the public goods aspect of street
order, especially in a large metropolitan city, where encounters are
anonymous and not necessarily repetitive as between the same 'players.'99
(Here we come back to the point about the insignificance of harms
inflicted by 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
traditional norms of conduct in public places now no longer apply, so far
as a significant number of citizens are concerned. I suspect that this is
widely recognized among Ellickson's pedestrians, and that it accounts in
part for their failure to participate enthusiastically in the informal social
enforcement exercise that he recommends. True - as I argued earlier101
- that recognition of the failure of complementarity is not particularly
cheerful. Often it is associated with anger, best explained in terms of
cognitive dissonance and denial. This leaves many people acutely
uncomfortable as they struggle to express, regulate, and make sense of
the annoyance that accrues from being confronted with the spectacle
96 J.Jacobs, TheDeath and Life of 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 Destruction of Social Capital Through Law' (1996)
144 U.Pa.L.Rev. 2055 at 2061 ff. and Livingston, supra note 54 at 558.
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, especially note 50 and accompanying text.

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
and the demands of street people. But the pedestrians that I see seem
more aware than Ellickson appears to be of the conditional nature of
street regulation. The regulation of public places is not like the categori-
cal imperative; it is sensitive to conditions and circumstances. I think
most people recognize that with the advent of large-scale homelessness
in our cities, conditions, and circumstances have changed; and, while this
may not necessarily excite a great deal of politically effective compassion,
still its implicit recognition may be enough to undermine - and (in an
indirect and no doubt back-handed Way) to appropriately undermine - the
sense of there being any acceptable ethical basis for the ordinary citizen
to chastise a panhandler or to roust and admonish a person who has
fallen asleep on the street.
I don't mean that normativity has disappeared altogether from the
streets. Quite the contrary: new norms may be emerging, or new selec-
tions being made in new circumstances from among the traditional social
rules. For example: many who give regularly to panhandlers understand
the importance of norms of time, place, and manner so far as begging
for change is concerned, and they do what they can to enforce these -
even if only by selective giving. (I know from conversation that many
people will contemplate giving only in circumstances where all parties to
the transaction have an opportunity to move away if they wish; they will
not, for example, under any circumstances, give money to a panhandler
in a crowded subway car.)102
Also, the more or less permanent presence of the homeless may
actually enhance street order in certain ways. Jane Jacobs spoke of the
importance of 'eyes upon the street'; she mentioned the special incen-
tives of merchants and regular users to observe and control what goes on
in a particular locale.10' Regular panhandlers have their eyes on the
street day and night, and often they know it as well as or better than its
more prosperous users. They often form bonds of affability, trust, even -
paradoxically - protectiveness vis-A-vis their regular alms-giving clientele.
And their sense of what is going on may well equal or even surpass -
occasionally they may be called on to supplement - the street knowledge
of an alert patrol officer.104
102 Ellickson's proposed test for this - 'How often has a New York commuter countered
a subway panhandler's monologue by, for example, starting a chant of "Just say
'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 Study in the Limits of Law' (1993) 27 Ind. L.R. 295 at 346,
that 'unaggressive chronic panhandlers may aid in police efforts to curb boisterous
transients,' Ellickson writes ('Controlling Chronic Misconduct,' supra note 12 at

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
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 every situation about which the indignant citizen
thinks, 'Something ought to be done.' The closeness to the ground (so to
speak) of the conditions under which social norms are constructed
means that they are likely to 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
manifestly unreasonable at the human level, then it may prove unen-
forceable, since its enforcement relies on the voluntary intervention of
those who are closely acquainted with the basis of the unreasonableness.
The contrast with the enforcement of official state law is important here.
In the case of a social norm, extra enforcement resources that are blind
to, or unaffected by, the sense of the norm's unreasonableness can
simply be wheeled in to supplant the reluctance of the enforcers on the
ground. (Social norms cannot mobilize task forces.) So, for example, if it
is well known that there are no public restrooms in a given vicinity, and
that homeless people have access to shelter facilities only at night, then
social norms are unlikely to spring up enforcing absolute prohibitions on
public urination. Instead, such social norms as there are on this issue will
likely direct street people to particular locations (in parks and public
gardens, etc.) where a modicum of privacy is available and where offence
to other members of the public is less likely.105 And the street-level
reasonableness of social norms in this regard is likely to be echoed also,
to a certain extent, in the exercise of ordinary discretion by police
officers on patrol. Someone urinating into the fountain in Lincoln Plaza
may be arrested; someone who has been seen behind a bush in Central
Park may not be.
These conclusions about social norms are significant for an evaluation
of the various legislative proposals that Ellickson and also the AARR put
forward. In Ellickson's presentation particularly, the proposals are put
forward as a legal response to the failure of social norms: it is the familiar
idea of law stepping into the breach where externalities and transaction
costs have defeated more informal processes. In fact, however, the im-
plications of the failure of social norms in this area may be more direct:
1198n), 'Most chronic panhandling, however, occurs in well-trafficked locations,

where there are likely to be numerous other eyes upon the street. A chronic
panhandler is therefore unlikely to make a net contribution to street order.'
105 Professor Ellickson and I were once given a tour of such sites in downtown Atlanta by
a homeless person (during a break in the conference mentioned supra note t). It was
made clear to us that, among street people, the designation of such sites was normative
as well as matter-of-fact - these were known as 'appropriate' places to, for example,
urinate - although I should add that I don't remember the word'normativity' actually
being used.

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
their lack of viability may provide useful information about the unreason-
ableness 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 against people who have little choice but to be in some
sense in violation of such norms.
XII Conclusion
The problem of homelessness has the potential to show communitarian-

ism at its best and at its worst.
At its best, communitarianism refuses to be browbeaten by either
conservative or liberal theories about what a good society must be like: it
encourages us to look to the conditions of viability of actualcommunities,
and it predicts that those conditions may sometimes surprise us, which
means that actually existing communities have the potential to teach us
something about social and political ideals. 'A community,' in Philip
Selznick's definition, 'is a comprehensive framework for social life.'106 It
provides a basis on which various human concerns are felt, values
pursued, and commitments entered into, among relatively large numbers
of people who accept that, since they are living more or less permanently
in one another's company, they have a responsibility to 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. the
certain sociological given-ness about community: it contrasts, on
hand, with utopia and social ideal, and, on the other hand, with
cult, and commune. One finds oneselfa member of a given comm
not by enlisting a group of like-minded chums, but by understa
who it is that one is 'unavoidably side-by-side' with, and who it is
impact of one's actions and decisions is going to be felt by, wheth
chooses this or not. The fellow members of one's community
necessarily people like oneself: they are, rather, those with who
cannot avoid interacting.'108 One is already in a community, alr
community with specified others; and one's communitarian resp
ties spring from that, not from one's wishes or ideals. Thus the f
someone smells bad, looks dishevelled, or is not the person on
choose to associate with does not mean that he is not a member of one's
community. If he is there, on the streets - the very streets that are the basis
of one's social, commercial, recreational interactions - then he is a
106 P. Selznick, 'The Idea of a Communitarian Morality' (1987) 75 Cal.L.Rev. 445 at 449.
107 See P. Selznick, The Moral Commonwealth (Berkeley: University of California Press,
1992) at 357-71.
108 These formulations are drawn from I. Kant, The Metaphysics ofMorals, trans. M. Gregor
(Cambridge: Cambridge University Press, 1991) at 121-4.

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
member of the community too. And any story one tells about communal
rights and responsibilities must take him and his interests into account.
But there is also a more disturbing side to contemporary com-
munitarianism. Communitarians often seem to be yearning for forms
and images of community that are no longer present or really possible.
They conjure up a delightful image of community, a sort of Norman
Rockwell picture of cheerful, prosperous people interacting respectfully
on the streets, in the parks, at church, in town meetings, at Little League
games, looking out for one another, and sustaining cherished 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
dirty or in despair, and there are no broken windows. I guess that as long
as these yearnings are confined to the realtor's brochures for actual or
notionally 'gated' communities, or to various Disneyfied experiments in
urban reconstruction,110 they are probably no more harmful than any
other cosmetic embodiment of American schwarmerei. But they become
dangerous, and they portend great injustice, when they are used as a
basis for determining rules of social conduct in community settings that
do not initially meet their specifications. When this happens, then,
although the resulting rules may be called 'community norms,' what they
109 See the extract from the AARR brief in Joyce, quoted at text accompanying note 9
supra.
110 The most explicit such experiment is the town of Celebration, Florida. See D. Young,
'The Laws of Community: The Normative Implications of Crime, Common Interest
Developments, and "Celebration"' (1998) 9 Hast.W.L.J. 121:
Celebration represents the privatization and increased corporate control of whole

aspects of the American lifestyle, a bastardization of the term 'community.' Much like
common interest developments, Celebration functions largely by conformity and
control rather than true cultural consensus and community decision-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 thing to representation in Celebration is membership in the
homeowner association. However, even the actions of this homeowner association can
be unilaterally overruled by the corporation. The sense of community so often
yearned for seems to be ironically misguided in the 'pretty calm' and 'civic infancy'
of Celebration. The longing for a bygone America, 'is a yearning for civic maturity'
in which the 'messy responsibility of democracy held sway, and society worked.'
Celebration does not offer these attributes.
See also A. Ross, The Celebration Chronicles: Life, Liberty and The Pursuit of Property Values
in Disney's New Towns (New York: Ballantine, 1999); M. Pollan, 'Town-Building Is No
Mickey Mouse Operation' The New York Times (14 December 1997) at 56; D. Tarrant,
'No Cause for Celebration: Disney-created Town 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' The New York Times (15 October 1998) Fl.

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC
actually represent is the imposition of a particular a priori vision by a few

persons (often outsiders) on a community that has traditionally orga-
nized itself on quite another basis.
A paragraph or two ago,"' I said that an attractive communitarianism
takes communities and their members as it finds them, not as its propo-
nents wish they would be. Those who have responsibility for the health of
the common environment in a particular place are those who actually
inhabit that place. This principle of the given-ness of community is quite
rightly invoked by Ellickson, Teir, and others when they argue that street
people too have responsibilities to the community - responsibilities, for
example, for the condition and safety of the community's public spaces.
Whether or not a homeless person has any choice about being on the
streets, the sheer fact of his being there means that he too has a duty to
the community in that regard. This we can accept. What we cannot
accept, however, is that the definition of communal responsibilities
should proceed on a basis that takes no account of the predicament of the
homeless person or of the particular nature of the stake that she may
have in the way public spaces are regulated. If the norms for public
spaces are to be observed by him, then the logic of genuine as opposed to
cosmetic communitarianism requires that those norms be constructed in
partforhim as well. We are not entitled to insist that the homeless person
abide by community norms, or that those norms be enforced against her,
if the norms are constructed in an image of community whose logic
denies in effect that homelessness exists.
In the end it comes down to a connection between community and
authenticity. In my particular criticisms of Ellickson, in sections VII and Ix
of this essay, I wanted to insist that the calculus of social costs and
benefits should not be used in a way that panders to the denial and the
fantasizing that distorts much American thinking about injustice. But the
broader message about homelessness and community as such is the
following. So long as people live among us in a condition of homeless-
ness, our normative definitions of community must be responsive to their
predicament; and it must be responsive, not only in articulating some
vague sense of social obligation to 'do something' about the problem,
but in accepting that the very definition of community must accommo-
date the stake that the homeless have - as community members - in the
regulation of public places. If the call for a greater emphasis on 'commu-
nity values' helps us see that, and helps us see our way through to new
and more hospitable conceptions of communal responsibility, then well
and good. But, as things stand, the call is most often heard in connection
with schemes of regulation that simply try to wish homeless members of
the community away. So long as that is the case, the moral credibility of
modern communitarianism remains a matter of doubt.
111 See text accompanying note 110 supra.

104.158.60.226 on Wed, 12 Jan 2022 05:05:44 UTC

Homelessness and Community

Uploaded by

Copyright:

Available Formats

Homelessness and Community

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Homelessness and Community

Uploaded by

Copyright:

Available Formats

Homelessness and Community

Author(s): Jeremy Waldron

Stable URL: https://www.jstor.org/stable/825960

This content downloaded from

I am not a communitarian, but I was prompted to ask questions about

(2000), 50 UNIVERSITY OF TORONTO LAWJOURNAL

This content downloaded from

community responsibility.2 The AARR filed an amicus brief supporting the

II Community control of public places

This content downloaded from

This content downloaded from

III Robert Ellickson

11 (1996) 105 Yale L.J. 1165 [hereinafter 'Controlling Chronic Misconduct'].

This content downloaded from

And Ellickson offers this definition of the phenomenon he takes as his

A person perpetrates a chronic street nuisance by persistently acting in a public

Unless chronic street nuisances of this sort are controlled, says

15 Ibid. at 1185 (emphasis omitted).

This content downloaded from

system of zoning in public places, permitting panhandling and other

IV Costs and benefits

What Ellickson's article and the AARR argument have in common is

This content downloaded from

considering the way in which he categorizes what he calls 'annoyance' in

Early in the article, Ellickson attempts what appears to be

Panhandlers and at least some of their donors benefit from b

For panhandlers, the benefit is the difference between be

[T]he magnitude depends on the quality of the bench squatter

22 'Controlling Chronic Misconduct,' supra note 12 at 1179.

24 Ibid. at 1183. (I make no comment on the plausibility - or tone - of this ana

This content downloaded from

This content downloaded from

v Are distress and annoyance harms?

The question I want to ask is whether the discomfort caused to ordinary

On the face of it, the question seems preposterous. Of course discom

This content downloaded from

VI The Appropriate Distress Argument

There is a well-known controversy about the application of the 'Harm

This content downloaded from

if one takes seriously the views one is propounding, and it is distressing to

Intemperance and the offence it inevitably occasions are indispens-

36 Mill, supra note 36 at 64-5.

This content downloaded from

Mill's utilitarian computations - for example, taking a long-run rather

The particular argument I want to make is that if Ellickson's pedestri-

39 'Utilitarianism' in J.S. Mill, Dissertations and Discussions: Political, Philosophical and

41 But compare R.K. Fullinwider, 'Patriotic History' in R.K. Fullinwider, ed., Pu

This content downloaded from

important that such repugnance not be treated as itself an evil on the

This content downloaded from

ment. As I emphasized in my earlier piece, the main argument for this

VII Cognitive dissonance

45 Waldron, 'Homelessness,' supra note 5 at 320-1 (Liberal Rights at 334-5).

This content downloaded from

VIII Broken windows

This content downloaded from

the view put forward by George Kelling andJames Q. Wilson in a famous

On this account, 'the ill-smelling drunk [or] the unchecked panhan-

Conception of Deterrence, the Broken Windows Theory, and Order-Maintenance

This content downloaded from