ON THE EXPRESSIVE FUNCTION OF LAW
CASS R. SUNSTEINt
IfI had known that not a single lunch counter would open as a result
of my action I could not have done differently than I did. If I had known
violence would result, I could not have done differently than I did. I am
thankfulfor the sit-ins iffor no other reason than that they provided me
with an opportunity for making a slogan into a reality, by turning a
decision into an action. It seems to me that this is what life is all about.
-Sandra Cason.1
We are all Expressionistspart of the time. Sometimes we just want to
scream loudly at injustice, or to stand up and be counted. These are noble
motives, but any serious revolutionist must often deprive himself of the
pleasures of self-expression. He must judge his actions by their ultimate
effects on institutions.
2
-Herbert Simon.
INTRODUCTION
Actions are expressive; they carry meanings.- This is true for
nearly everything we do, from the most mundane to the most
significant. For example, a lawyer who wears a loud tie to court will
be signalling something distinctive about his self-conception and his
attitude toward others; so too with a law professor who teaches in
blue jeans; so too with a student who comes to class in a business
suit. What can be said for nonverbal acts applies to purely verbal
statements as well. A bank president who uses the terms "Miss" and
"Mrs.," or who refers to African Americans as "Negroes," will be
showing a wide range of things about his attitudes on matters of
t Karl N. Llewellyn Distinguished Service Professor ofJurisprudence, University
of Chicago. I am grateful to Dan Kahan, Lawrence Lessig, Martha Nussbaum,
Richard Posner, and David Strauss for valuable comments on an earlier draft.
'JAMES MILLER, "DEMOCRACY IS IN THE STREETS": FROM PORT HURON TO THE
SIEGE
2
OF CHICAGO 52 (1987).
HERBERT A. SIMON, MODELS OF MY LIFE 281 (1991).
- In law, Lawrence Lessig offers the best discussion of this idea in The Regulation
of Social Meaning, 62 U. CHI. L. REV. 943 (1995). I owe a general debt to Lessig's
important paper and especially to his discussion of the collective action problem
posed by efforts at changing social meanings. See id. at 993-1007.
(2021)
,2022 UNIVERSITY OF PENNSYLVANIA LAW REVIEW [Vol. 144: 2021
gender and race. So too with a Southern politician who uses the
terms "Ms." and "African American."
In these and other cases, what the agent will be communicating,
or be taken to mean, may or may not have a great deal to do with
his particular intentions. In this sense, the meanings of actions are
not fully within the agent's control. Indeed, some agents may not
even be aware of the relevant meanings. Consider a foreigner
whose very foreignness is often signalled by obliviousness to the
social meanings of his actions. What he says may be very different
from what he means.
The social meanings of actions are very much a function of
existing social norms. When a social norm tells people not to
smoke in public places, the social meaning of smoking is obtuseness,
discourtesy, or worse. When a social norm requires people to dress
casually for dinner, formal attire "means" something bad, like a
desire to seem superior or a manifestation of an odd social rigidity.
And when social norms change, social meaning changes too. Thus
the social meanings of lighting up a cigarette, or engaging in an act
of sexual harassment, or using a condom, or refusing to eat meat,
are very different in 1996 from what they were in 1966, because of
dramatic shifts in underlying norms.
What can be said for actions can also be said for law. Many
people support law because of the statements made by law, and
disagreements about law are frequently debates over the expressive
content of law. Much of the debate over school segregation, for
example, was also a debate about the meaning of laws calling for
segregation. Plessy v. Ferguson4 asserted that such laws did not
"mean" black inferiority;5 Brown v. Board of Education6 tried to
respond to this assertion with empirical work suggesting the
contrary.' Or, consider debates over capital punishment. Many
people who oppose capital punishment would be unlikely to shift
their position even if evidence were to show that capital punishment
does have a deterrent effect. They are concerned about the
expressive content of capital punishment, not about its ineffective-
ness as a deterrent (or about other nonexpressive grounds for
4 163 U.S. 537 (1896).
' See id. at 544 (noting that segregation laws "do not necessarily imply the
inferiority of either race to the other").
6 347 U.S. 483 (1954).
See id. at 494 n.1 1 (relying on social science evidence that showed that segrega-
tion generated a feeling of inferiority among African-Americans).
, ON THE EXPRESSIVE FUNCTION OF LAW 2023
punishment).8 And many people who endorse capital punishment
would not be much moved by evidence that capital punishment does
not deter people from committing crimes. Their primary concern
is the symbolic9
or expressive content of the law, not aggregate
murder rates.
Very recently, the enormously lengthy and heated debate over
flag burning has been permeated by expressive concerns. If we ask
whether the debate is about how best to deter flag burning, we will
find the debate unintelligible. Few people have burned the
American flag in recent years, and it is reasonable to suppose that
a constitutional amendment making it possible to criminalize flag
burning would have among its principal consequences a dramatic
increase in annual acts of flag burning. In fact, adopting a
constitutional amendment may be the best possible way to promote
the incidence of flag burning. In these circumstances it seems clear
that those who support the amendment are motivated not so much
by consequences as by expressive concerns.1 0 They appear to want
to make a statement about the venality of the act of flag burning,
perhaps in order to affect social norms, perhaps because they think
that making the statement is intrinsically good.
Much of the contemporary debate over the regulation of hate
speech is similar. It is above all about the social meaning of such
regulations. Do such regulations "mean" that victims of hate speech
require special paternalistic protections, are weak and thin-skinned,
and unable to take care of themselves? Or do they "mean" that
bigotry is utterly unacceptable in a liberal society? Debates of this
kind could not plausibly be focused on consequences, for the stakes
are relatively low and thus cannot justify the amount of time and
energy devoted to the issue. In this way, debates over flag burning
8 1 do not attempt here to sort out the relation between expressive and other
grounds for criminal punishment.
9
See Tom R. Tyler & Renee Weber, Supportfor the Death Penalty: Instrumental
Response to Crime, or Symbolic Attitude?, 17 L. &SOC'Y REV. 21, 40 (1982) (finding that
"[p]olitical and social beliefs ... exercise[d] a strong influence upon support for
capital punishment, while the influence of crime-related concerns was small"); Phoebe
C. Ellsworth & Samuel R. Gross, Hardeningof the Attitudes: Americans' Views on the
Death Penalty,J.Soc. ISSUES, Summer 1994, at 19,27 (finding that, while belief in the
death penalty as an effective deterrent hovered around 60% from 1972 to 1991,
support for the death penalty during that period rose from 58% to 75%).
10The same thing might be said about any law that drives conduct underground.
Even if the aggregate incidence of harmful conduct is not decreased, some people
support relevant laws on expressive grounds. Consider prohibitions on the use of
drugs or on certain sexual activity such as homosexuality, adultery, and fornication.
, 2024 UNIVERSITY OF PENNSYLVANIA LAW REVIEW [Vol. 144: 2021
and debates over hate speech have a great deal in common; they are
expressive in character.
Consider, too, the subject of risk regulation. In environmental
protection, public debate is often focused on the perceived social
meaning of law. Thus the Endangered Species Act has a special
salience as a symbol of a certain conception of the relationship
between human beings and their environment, and emissions
trading systems are frequently challenged because they are said to
"make a statement" that reflects an inappropriate valuation of the
environment. 1 In the same way, mandatory recycling (as opposed
to curbside charges, which seem far better from an economic
standpoint 2 ) may well receive public support on expressive grounds.
In the legal profession, the same may also be true of mandatory pro
bono work (as opposed to compulsory donations from lawyers who
refuse to do such work)."3
In this Article I explore the expressive function of law-the
function of law in "making statements" as opposed to controlling
behavior directly. 4 I do so by focusing on the particular issue of
"See STEVEN KELMAN, WHAT PRICE INCENTIVES?: ECONOMISTS AND THE
ENVIRONMENT 2 (1981) ("[T]he question of whether or not to use economic incentives
in environmental policy is not simply a technical question, but is also an ideological,
philosophical question.., and many noneconomist participants in the environmental
debate tend to react to the issue in ideological terms."). I do not mean to endorse
the view that emissions trading systems reflect an inappropriate valuation of the
environment.
12 See Peter S. Menell, Beyond the Throwaway Society: An Incentive Approach to
Regulating Municipal Solid Waste, 17 ECOLOGY L.Q. 655, 696 (1990) (comparing
various solid waste regulatory policies and finding that on a purely economic basis
perfect curbside charge policy and perfect deposit-refund policy always achieve the
first-best allocation of resources).
" See Richard H. Pildes, The Unintended Cultural Consequences of PublicPolicy: A
Comment on the Symposium, 89 MICH. L. REV. 936, 947-51 (1991) (arguing that
mandatory pro bono requirements will not affect lawyers' conception of their
professional duty if they are allowed to hire others to do the work on their behalf).
14 For valuable discussions of the expressive function of legal and economic
norms, see ELIZABETH ANDERSON, VALUE IN ETHICS AND ECONOMICS 33-37 (1993)
(discussing expressive norms); ROBERT NOZICK, THE NATURE OF RATIONALITY 26-35
(1993) [hereinafter NOZICK, RATIONALITY] (discussing "symbolic utility," the utility
that is imputed to an action or outcome in accordance with its symbolic meaning);
ROBERT NOZICK, PHILOSOPHICAL EXPLANATIONS 370-88 (1981) [hereinafter NOZICK,
EXPLANATIONS] (discussing the symbolic message of retributive punishment); Jean
Hampton, An Expressive Theoy of Retribution, in RETRIBUTIVISM AND ITS CRITICS 1
(Wesley Cragg ed. 1992) (developing an expressive theory of retribution to explain
and justify retributive practice); see also Dan M. Kahan, What Do Alternative Sanctions
Mean?, 63 U. CHI. L. REV. (forthcoming Spring 1996) (manuscript at 3) ("Punishment
is notjust a way to make offenders suffer; it is a special social convention that signifies
moral condemnation."); Richard H. Pildes & Cass R. Sunstein, Reinventing the
CASS R. SUNSTEINt
IfI had known that not a single lunch counter would open as a result
of my action I could not have done differently than I did. If I had known
violence would result, I could not have done differently than I did. I am
thankfulfor the sit-ins iffor no other reason than that they provided me
with an opportunity for making a slogan into a reality, by turning a
decision into an action. It seems to me that this is what life is all about.
-Sandra Cason.1
We are all Expressionistspart of the time. Sometimes we just want to
scream loudly at injustice, or to stand up and be counted. These are noble
motives, but any serious revolutionist must often deprive himself of the
pleasures of self-expression. He must judge his actions by their ultimate
effects on institutions.
2
-Herbert Simon.
INTRODUCTION
Actions are expressive; they carry meanings.- This is true for
nearly everything we do, from the most mundane to the most
significant. For example, a lawyer who wears a loud tie to court will
be signalling something distinctive about his self-conception and his
attitude toward others; so too with a law professor who teaches in
blue jeans; so too with a student who comes to class in a business
suit. What can be said for nonverbal acts applies to purely verbal
statements as well. A bank president who uses the terms "Miss" and
"Mrs.," or who refers to African Americans as "Negroes," will be
showing a wide range of things about his attitudes on matters of
t Karl N. Llewellyn Distinguished Service Professor ofJurisprudence, University
of Chicago. I am grateful to Dan Kahan, Lawrence Lessig, Martha Nussbaum,
Richard Posner, and David Strauss for valuable comments on an earlier draft.
'JAMES MILLER, "DEMOCRACY IS IN THE STREETS": FROM PORT HURON TO THE
SIEGE
2
OF CHICAGO 52 (1987).
HERBERT A. SIMON, MODELS OF MY LIFE 281 (1991).
- In law, Lawrence Lessig offers the best discussion of this idea in The Regulation
of Social Meaning, 62 U. CHI. L. REV. 943 (1995). I owe a general debt to Lessig's
important paper and especially to his discussion of the collective action problem
posed by efforts at changing social meanings. See id. at 993-1007.
(2021)
,2022 UNIVERSITY OF PENNSYLVANIA LAW REVIEW [Vol. 144: 2021
gender and race. So too with a Southern politician who uses the
terms "Ms." and "African American."
In these and other cases, what the agent will be communicating,
or be taken to mean, may or may not have a great deal to do with
his particular intentions. In this sense, the meanings of actions are
not fully within the agent's control. Indeed, some agents may not
even be aware of the relevant meanings. Consider a foreigner
whose very foreignness is often signalled by obliviousness to the
social meanings of his actions. What he says may be very different
from what he means.
The social meanings of actions are very much a function of
existing social norms. When a social norm tells people not to
smoke in public places, the social meaning of smoking is obtuseness,
discourtesy, or worse. When a social norm requires people to dress
casually for dinner, formal attire "means" something bad, like a
desire to seem superior or a manifestation of an odd social rigidity.
And when social norms change, social meaning changes too. Thus
the social meanings of lighting up a cigarette, or engaging in an act
of sexual harassment, or using a condom, or refusing to eat meat,
are very different in 1996 from what they were in 1966, because of
dramatic shifts in underlying norms.
What can be said for actions can also be said for law. Many
people support law because of the statements made by law, and
disagreements about law are frequently debates over the expressive
content of law. Much of the debate over school segregation, for
example, was also a debate about the meaning of laws calling for
segregation. Plessy v. Ferguson4 asserted that such laws did not
"mean" black inferiority;5 Brown v. Board of Education6 tried to
respond to this assertion with empirical work suggesting the
contrary.' Or, consider debates over capital punishment. Many
people who oppose capital punishment would be unlikely to shift
their position even if evidence were to show that capital punishment
does have a deterrent effect. They are concerned about the
expressive content of capital punishment, not about its ineffective-
ness as a deterrent (or about other nonexpressive grounds for
4 163 U.S. 537 (1896).
' See id. at 544 (noting that segregation laws "do not necessarily imply the
inferiority of either race to the other").
6 347 U.S. 483 (1954).
See id. at 494 n.1 1 (relying on social science evidence that showed that segrega-
tion generated a feeling of inferiority among African-Americans).
, ON THE EXPRESSIVE FUNCTION OF LAW 2023
punishment).8 And many people who endorse capital punishment
would not be much moved by evidence that capital punishment does
not deter people from committing crimes. Their primary concern
is the symbolic9
or expressive content of the law, not aggregate
murder rates.
Very recently, the enormously lengthy and heated debate over
flag burning has been permeated by expressive concerns. If we ask
whether the debate is about how best to deter flag burning, we will
find the debate unintelligible. Few people have burned the
American flag in recent years, and it is reasonable to suppose that
a constitutional amendment making it possible to criminalize flag
burning would have among its principal consequences a dramatic
increase in annual acts of flag burning. In fact, adopting a
constitutional amendment may be the best possible way to promote
the incidence of flag burning. In these circumstances it seems clear
that those who support the amendment are motivated not so much
by consequences as by expressive concerns.1 0 They appear to want
to make a statement about the venality of the act of flag burning,
perhaps in order to affect social norms, perhaps because they think
that making the statement is intrinsically good.
Much of the contemporary debate over the regulation of hate
speech is similar. It is above all about the social meaning of such
regulations. Do such regulations "mean" that victims of hate speech
require special paternalistic protections, are weak and thin-skinned,
and unable to take care of themselves? Or do they "mean" that
bigotry is utterly unacceptable in a liberal society? Debates of this
kind could not plausibly be focused on consequences, for the stakes
are relatively low and thus cannot justify the amount of time and
energy devoted to the issue. In this way, debates over flag burning
8 1 do not attempt here to sort out the relation between expressive and other
grounds for criminal punishment.
9
See Tom R. Tyler & Renee Weber, Supportfor the Death Penalty: Instrumental
Response to Crime, or Symbolic Attitude?, 17 L. &SOC'Y REV. 21, 40 (1982) (finding that
"[p]olitical and social beliefs ... exercise[d] a strong influence upon support for
capital punishment, while the influence of crime-related concerns was small"); Phoebe
C. Ellsworth & Samuel R. Gross, Hardeningof the Attitudes: Americans' Views on the
Death Penalty,J.Soc. ISSUES, Summer 1994, at 19,27 (finding that, while belief in the
death penalty as an effective deterrent hovered around 60% from 1972 to 1991,
support for the death penalty during that period rose from 58% to 75%).
10The same thing might be said about any law that drives conduct underground.
Even if the aggregate incidence of harmful conduct is not decreased, some people
support relevant laws on expressive grounds. Consider prohibitions on the use of
drugs or on certain sexual activity such as homosexuality, adultery, and fornication.
, 2024 UNIVERSITY OF PENNSYLVANIA LAW REVIEW [Vol. 144: 2021
and debates over hate speech have a great deal in common; they are
expressive in character.
Consider, too, the subject of risk regulation. In environmental
protection, public debate is often focused on the perceived social
meaning of law. Thus the Endangered Species Act has a special
salience as a symbol of a certain conception of the relationship
between human beings and their environment, and emissions
trading systems are frequently challenged because they are said to
"make a statement" that reflects an inappropriate valuation of the
environment. 1 In the same way, mandatory recycling (as opposed
to curbside charges, which seem far better from an economic
standpoint 2 ) may well receive public support on expressive grounds.
In the legal profession, the same may also be true of mandatory pro
bono work (as opposed to compulsory donations from lawyers who
refuse to do such work)."3
In this Article I explore the expressive function of law-the
function of law in "making statements" as opposed to controlling
behavior directly. 4 I do so by focusing on the particular issue of
"See STEVEN KELMAN, WHAT PRICE INCENTIVES?: ECONOMISTS AND THE
ENVIRONMENT 2 (1981) ("[T]he question of whether or not to use economic incentives
in environmental policy is not simply a technical question, but is also an ideological,
philosophical question.., and many noneconomist participants in the environmental
debate tend to react to the issue in ideological terms."). I do not mean to endorse
the view that emissions trading systems reflect an inappropriate valuation of the
environment.
12 See Peter S. Menell, Beyond the Throwaway Society: An Incentive Approach to
Regulating Municipal Solid Waste, 17 ECOLOGY L.Q. 655, 696 (1990) (comparing
various solid waste regulatory policies and finding that on a purely economic basis
perfect curbside charge policy and perfect deposit-refund policy always achieve the
first-best allocation of resources).
" See Richard H. Pildes, The Unintended Cultural Consequences of PublicPolicy: A
Comment on the Symposium, 89 MICH. L. REV. 936, 947-51 (1991) (arguing that
mandatory pro bono requirements will not affect lawyers' conception of their
professional duty if they are allowed to hire others to do the work on their behalf).
14 For valuable discussions of the expressive function of legal and economic
norms, see ELIZABETH ANDERSON, VALUE IN ETHICS AND ECONOMICS 33-37 (1993)
(discussing expressive norms); ROBERT NOZICK, THE NATURE OF RATIONALITY 26-35
(1993) [hereinafter NOZICK, RATIONALITY] (discussing "symbolic utility," the utility
that is imputed to an action or outcome in accordance with its symbolic meaning);
ROBERT NOZICK, PHILOSOPHICAL EXPLANATIONS 370-88 (1981) [hereinafter NOZICK,
EXPLANATIONS] (discussing the symbolic message of retributive punishment); Jean
Hampton, An Expressive Theoy of Retribution, in RETRIBUTIVISM AND ITS CRITICS 1
(Wesley Cragg ed. 1992) (developing an expressive theory of retribution to explain
and justify retributive practice); see also Dan M. Kahan, What Do Alternative Sanctions
Mean?, 63 U. CHI. L. REV. (forthcoming Spring 1996) (manuscript at 3) ("Punishment
is notjust a way to make offenders suffer; it is a special social convention that signifies
moral condemnation."); Richard H. Pildes & Cass R. Sunstein, Reinventing the
