TTC - Law School For Everyone, Legislation and Regulation

Topic Subtopic
Professional Law
Law School
for Everyone
Legislation and Regulation
Course Guidebook
Professor Peter J. Smith

The George Washington University Law School
PUBLISHED BY:
THE GREAT COURSES
Corporate Headquarters
4840 Westfields Boulevard Suite 500

Chantilly, Virginia 20151‑2299
Phone 1.800.832.2412 | Fax 703.378.3819
www.thegreatcourses.com
Copyright © The Teaching Company, 2019
Printed in the United States of America

This book is in copyright. All rights reserved.
Without limiting the rights under copyright reserved above, no part of this publication
may be reproduced, stored in or introduced into a retrieval system, or transmitted,
in any form, or by any means (electronic, mechanical, photocopying, recording, or
otherwise), without the prior written permission of The Teaching Company.
Peter J. Smith, JD
Arthur Selwyn Miller

Research Professor of Law
The George Washington

University Law School
P
eter J. Smith is the Arthur Selwyn
Miller Research Professor of
Law at The George Washington
University Law School in Washington
DC. He received his BA from Yale
University, where he graduated magna cum laude, and his JD from Harvard
Law School, where he graduated magna cum laude and received the Sears
Prize for highest academic performance.
Before joining the faculty at GW Law, Professor Smith was an attorney at

the US Department of Justice, where he represented the government in
the US Courts of Appeals. At the Department of Justice, he defended the
constitutionality of a number of federal statutes, including the Family and
Medical Leave Act and the Food and Drug Administration Modernization
Act, in cases that ultimately were resolved by the Supreme Court. Before he
worked at the Department of Justice, Professor Smith clerked for Judge Phyllis
A. Kravitch of the US Court of Appeals for the Eleventh Circuit.
Professor Biography I
Professor Smith has twice received the Distinguished Faculty Service Award
for outstanding teaching at GW Law. He has published dozens of scholarly
articles and is the coauthor of a popular casebook on constitutional law titled
Constitutional Law: A Contemporary Approach. His research focuses on
constitutional law, constitutional interpretation, and civil procedure.
Professor Smith also teaches the civil procedure section of the Great Course
Law School for Everyone. ■
II Law School for Everyone: Legislation and Regulation

TABLE OF CONTENTS
Introduction
Professor Biography . . . . . . . . . . . . . . . . . . . . . . . . . . . . I
Table of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . III
Course Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Lecture Guides
1• Making Sense of Legislation and Regulation . . . . . . . . . . . . . . . . 3
2• Regulation by Statute and by Common Law . . . . . . . . . . . . . . . 10
3• Legislation and the Administrative State . . . . . . . . . . . . . . . . . . 19
4• Touchstones of Statutory Interpretation . . . . . . . . . . . . . . . . . . 29
5• The Letter versus the Spirit of the Law . . . . . . . . . . . . . . . . . . . . 35
6• When Is Statutory Meaning Plain? . . . . . . . . . . . . . . . . . . . . . . . . 42
7• Semantic and Substantive Interpretive Rules . . . . . . . . . . . . . . 50
8• How Do Courts Really Interpret Statutes? . . . . . . . . . . . . . . . . . 58
9• Federal Agencies as Regulatory Bodies . . . . . . . . . . . . . . . . . . . 68
10• Political Control of Agency Decision Making . . . . . . . . . . . . . . . 77
11• Judicial Review of Agency Rulings . . . . . . . . . . . . . . . . . . . . . . . 86
12• Weighing Agency Interpretations of Statutes . . . . . . . . . . . . . . 95
Supplementary Material
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
Image Credits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
Table of Contents III

DISCLAIMER
The legal information provided in these lectures is for informational purposes
only and not for the purpose of providing legal advice. These lectures may
not reflect the most current legal developments in any particular applicable
jurisdictions and cannot substitute for the advice of a licensed professional
with specialized knowledge who can apply it to the particular circumstances
of your situation. Use of and access to these lectures do not create an attorney-
client relationship with The Teaching Company or its lecturers, and neither
The Teaching Company nor the lecturer is responsible for your use of this
educational material or its consequences. You should contact an attorney
to obtain advice with respect to any particular legal issue or problem. The
opinions and positions provided in these lectures reflect the opinions and
positions of the relevant lecturer and do not necessarily reflect the opinions or
positions of The Teaching Company or its affiliates. Pursuant to IRS Circular
230, any tax advice provided in these lectures may not be used to avoid tax
penalties or to promote, market, or recommend any matter therein.
The Teaching Company expressly DISCLAIMS LIABILITY for any DIRECT,

INDIRECT, INCIDENTAL, SPECIAL, OR CONSEQUENTIAL DAMAGES
OR LOST PROFITS that result directly or indirectly from the use of these
lectures. In states that do not allow some or all of the above limitations of
liability, liability shall be limited to the greatest extent allowed by law.
IV Law School for Everyone: Legislation and Regulation

LAW SCHOOL FOR EVERYONE
LEGISLATION AND
REGULATION
I
n law school, a course on legislation and regulation is about the rules that
govern the enactment, interpretation, and implementation of statutes. Such
a course is different from other first-year courses in law school. Most of
the other courses taught in the first year of law school—on torts, property,
and contracts, for example—are common-law subjects. Under the common
law, courts devise rules to resolve the cases presented to them, and they write
careful opinions to explain the rules and justify the conclusions that they reach.
However, the main texts in a course on legislation and regulation are statutes
enacted by legislatures and regulations issued by agencies. Whereas judicial
opinions announce rules and reasons, statutes and regulations tend to
announce just rules. If judicial opinions are novels, then statutes are poems:
They are brief, yet packed with meaning. A course on legislation and regulation
is in large part about how to discern that meaning.
This course begins by considering a deceptively simple statute banning

vehicles in the park. It turns out that statutory interpretation is much more
complicated than it might initially seem. We will also consider the justifications
for government regulation and the reasons why we might choose to regulate
by statute rather than by the common-law system.
Course Scope 1
After an overview of the legislative process, we will turn to the complex world
of statutory interpretation. Important considerations arise: Should courts
enforce the plain meaning of statutes, the legislature’s intent or purpose, or
the judges’ sense of policy and fairness? We will address these questions by
considering the various touchstones of interpretation, with lectures focused
on the letter and spirit of the law, the canons of interpretation, and theories
of the courts’ role in statutory interpretation.
The course then turns to the administrative state. We will consider where
federal agencies get their power to regulate and how exactly they use it. We
will start by focusing on the nondelegation doctrine, which governs the limits
on Congress’s power to give away its policy-making power. Then, we will
consider political and judicial controls on agency power. The course ends with
a lecture on the distinctive questions about statutory interpretation that arise
in the administrative state. ■
2 Law School for Everyone: Legislation and Regulation

Lecture 1
MAKING SENSE OF
LEGISLATION AND
REGULATION
I
t is in areas of ambiguity that lawyers often make their greatest contributions.
Persuading a court that one’s interpretation of a statute or a regulation
is correct can have an enormous impact: It can heal a grave injustice, or
significantly change a governmental policy, or open an entire new realm of
opportunity for science or industry. And it all may turn on how lawyers and
judges parse a few words on a page.
This is the fascinating world of legislation and regulation. To introduce you to it,

this lecture covers some background information and hypothetical stories.
SETTING THE STAGE
Imagine that your 15-year-old son just got home. He is upset and angry
because a police officer stopped him while he was riding his bicycle
through the park. He was on the way home from school, and he
kept to the paved paths in the park. But the police officer gave him
a citation, saying that your son violated a town ordinance. The
ordinance states, “No vehicles shall be permitted in the park.”
Has your son actually broken the law? That depends

on what the law means by the term vehicles.
Is a bicycle a vehicle? If it is, then it seems
like your son has unfortunately violated
the ordinance. But is it obvious that
a bicycle is a vehicle?
If you live in a dictatorship, where the government can punish you at its whim,
then that is not the kind of question you are likely to ask. However, in countries
like the United States, where the government is supposed to be subject to the
rule of law, it matters what the law says. If the law does not ban bicycles from
the park, then your son cannot be punished for riding one in the park.
But how are we supposed to know what the law prohibits? Sometimes this is
a challenge because we are simply unaware of the law. Even if ignorance was
not at play in our example, imagine there was a big sign posted outside the
park that clearly said, “No Vehicles Shall Be Permitted in the Park.” We still
need to interpret the law to figure out whether it bans bicycles.
Interpreting statutes and regulations, and understanding the rules by which

the government adopts them and courts apply them, is the subject of a typical
first-year law school course on legislation and regulation. This is a relatively
recent addition to the first-year curriculum. Virtually all the other courses
taught in the first year—torts, property, and contracts, for example—are
primarily common-law subjects.
Under the common-law system, which the United States inherited from
England, courts devise rules to resolve the cases presented to them, and they
look to the reasoning in opinions from previous cases for guidance on how
to decide new controversies. It is these judicial opinions that students of torts,
property, and contracts spend much of their time reading and discussing.

However, the common-law process is no longer the central feature of our
legal system. Instead, law today is more likely to be enacted by a legislature or
promulgated by an administrative agency than it is to be declared by a court.
That does not mean it has any less authority, however. If you want to figure
out how to comply with a statute or a regulation, or challenge one, or help
a client do the same, you need to know how to make sense of them.
IS IT A VEHICLE?
To return to your son’s predicament, it is pointless to fume about the citation

that the police officer gave him. It is more productive to figure out whether
a bicycle is vehicle. One option is to look in the dictionary, but in some cases
this may not answer the question, especially when multiple definitions of
vehicle or ones that do not seem to fit the term bicycle are in play.
Perhaps it is more useful answer the question about the law’s meaning by
referring to the word’s ordinary, commonsense meaning. For example, when
you hear the term vehicle, do you think of bicycles? Most people do not.
However, what if instead of your son, the citation was for an adult courier
who knocked a pedestrian down? It would not be as obvious, in that case, that
there had not been a violation of the ordinance.
Because the rule in question is a town ordinance, another way to look at it is

what the town council meant when it used the word vehicle in the ordinance.
Perhaps their intent was to ban bicycles from the town’s parks.
These approaches highlight an important difference between courses that

cover common-law subjects and courses that focus on statutes and regulations.
Students in subjects like torts, contracts, and property learn to read judicial
decisions and then to make arguments about why the reasoning in a prior
case mandates a particular result in a new case.
However, statutes usually do not include reasoning—that is, they do not

ordinarily explain why the legislature addressed a particular problem or what
the justification for the regulation was. Whereas judicial opinions announce
rules and reasons, statutes tend to just announce rules.
Lecture 1–Making Sense of Legislation and Regulation 5

THE AMBIGUITY OF LANGUAGE
Another feature of interpreting legislation and regulation is that the ambiguity

of language frequently comes into play. Indeed, the indeterminacy of language
is endemic in the law. The Constitution, for example, protects the “freedom
of speech.” Does this concept embrace the right to burn the American flag as
a form of political protest? Does it protect the right of a corporation to spend
money to promote the election of a candidate for office?
The problem arises even with more mundane statutes that, at least at first
blush, appear to speak with clarity. For example, suppose a state law requires
drivers to turn on their headlights “after dark.” Does a driver violate the rule
by failing to turn on his lights at dusk? At some indefinable point, it transitions
from being light out to being dark out. How do we decide how the law applies
in the face of this sort of ambiguity?
A legislature could take care of this problem by creating a bright-line rule

that is easy to administer. For example, the law could state that drivers must
“use their headlights 30 minutes before sunset until 30 minutes after sunrise.”

Then, we would just have to check our weather app to know what time sunrise
and sunset occurred, and that would answer whether headlights in fact were
required. However, there would still be at least two problems with this law.
First, most people don’t know exactly when sunrise and sunset are on a given
day. Second, we would still want a provision that says that drivers also need
to use their headlights at other times of day if it happens, because of weather
conditions, to be unusually dark out. Then we would be right back to our
interpretive problem.
LETTER AND SPIRIT
Another complication arises when the plain terms of the law conflict with
its ostensible spirit.
Consider this example: Imagine that the town owns and operates the
local train station. A town ordinance provides, “Sleeping in the train
station is a misdemeanor punishable by a fine of not more than $500 or
imprisonment of no more than two days.” A frequent business traveler
arrives at the train station to learn that his train has been delayed an hour.
He sits down in the waiting area and, after a few minutes, he dozes off. Has
he violated the law?
The ordinance clearly prohibits sleeping in the station, with no listed

exceptions. And he was, in fact, sleeping in the station. Both the ordinance
and the facts seem clear. But would it not be unjust to punish him?
The law likely was designed to prevent vagrants from living in the station,
and the facts of our case are sufficiently different from that targeted
activity that it would seem inconsistent with the spirit of the law to punish
him. Even with this reasoning, however, nailing down the spirit of a law can
be difficult.
This complication also raises questions about the role of judges. A simple
way to describe their role is that judges interpret laws. Laws come from
democratically elected representatives.

In one view, judges ought to exercise judgment in deciding when a particular
application of a law would be unjust or even ridiculous. In another view, if
judges can ignore the law and replace it with their personal views about fairness
or wise policy, then are acting like legislators. But in most places, we do not
get to vote for the judges.
This raises questions: Is it the judge’s role to interpret a statute to avoid

troubling, ridiculous, or unfair results? Or is it problematic for judges, many
of whom are not elected, to substitute their views about what is appropriate
for those of our elected representatives? This, of course, is a matter of intense
political controversy in American law and government.
OTHER INTERPRETATIVE QUESTIONS
Returning to the no-vehicles-in-the-park example, assume that its meaning

is relatively clear with respect to most modes of transportation that you can
use today. What would happen if, in 20 years, a teenager brought
his brand-new hovercraft to the park? Imagine that the craft
would permit him to travel through the park 15 feet off the
ground, with its silent engine barely detectable to park goers.
If the ordinance hasn’t changed—it still says, “No vehicles
shall be permitted in the park”—then is the teenager in
violation of the law?
To answer this question, we would not only need to define

the term vehicle; we would also need to decide at what level
of generality to understand the term. After all, if the term
applies only to vehicles that existed at the time the ordinance
was enacted, then the legislature would have to update the
law every time the world changed—which is constantly.
But if the term can embrace unforeseen circumstances,

then we are confronted with the additional challenge
of deciding what types of new vehicles are sufficiently
similar to older vehicles that they should count. The US Supreme Court often
confronts these kinds of questions, and there are often lively debates among
the justices about how to resolve them.

Another complication is administrative agencies. Imagine that the town’s
department of parks and recreation recently issued guidance to park officials
and police, instructing them to apply the ban to bicycles and skateboards,
but not to roller skates, tricycles, or baby carriages. Should the agency’s
understanding of the park law influence the court’s view about what it means?
To answer such questions, it is necessary to learn more about the role of

administrative agencies in our system. It is difficult, if not impossible, to
fully resolve a question of statutory interpretation without understanding
the nature of regulation in our administrative state.
Suggested Reading
<< Dickerson, The Interpretation and Application
of Statutes.
<< Popkin, Statutes in Court.
<< Scalia, A Matter of Interpretation.
Questions to Consider
ÞÞ How do we determine the meaning of a statute?
ÞÞ Is legal meaning ever really plain?

Lecture 2
REGULATION BY STATUTE
AND BY COMMON LAW
D
iscussions about regulation carry all sorts of political baggage, but
the term regulate simply means to “control or supervise by means of
rules,” and regulation refers to a set of rules made and maintained by
some government authority. Yet, where regulation is involved, things are rarely
simple, as this lecture shows through how both the courts and Congress have
tackled particularly difficult questions.
ENVIRONMENTAL REGULATION
Environmental regulation is an interesting case study. For example, questions

arise such as: Should the government limit the amount of carbon dioxide
and other greenhouse gases that factories can emit, or prohibit the sale of
gas-guzzling cars to reduce the effects of climate change? If the government
chooses to do so, how should it decide just how much carbon dioxide is too
much, or just how fuel efficient new cars must be?
These questions do not have easy answers. Part of the difficulty has to do
with the limits on our ability to predict the effects of regulation. For example,
will strict fuel-efficiency standards lead
to innovation and clean air without
substantially raising the price of CRIMINAL LAW
cars? Alternatively, will they threaten
the solvency of car manufacturers Though much regulation is
and limit the ability of the poor and controversial, virtually all people
middle class to afford a crucial means can agree that the government
of transportation? properly prohibits murder and
theft, thereby controlling our
In some cases, it is likely that regulation behavior by means of rules.
is not justified. However, it is fair to say Criminal law, in other words, is
that at least sometimes, there will be a form of regulation.
a need for some form of government
regulation.
CONTRACTS AND TORT LAW
We might have trouble reaching a universal agreement about the need for
regulation to limit the use of coal, but there probably are not many people
who think that the government shouldn’t enforce contracts between private
parties. Courts enforce most contracts to ensure that parties are held to
promises that others have relied upon. As a consequence, contracting parties
generally conform their conduct to the rules that courts have announced as
a matter of contract law, and when they do not, they can be ordered by a court
to pay damages.
Lecture 2–Regulation by Statute and by Common Law 11

The same is true about tort law. Tort law concerns the law of accidents; it is
how our legal system decides when one person should be liable for injuries
suffered by another person.
Tort law and contract law are classic common-law subjects. The common
law is a system in which courts devise rules to resolve the cases presented
to them. Under that system, when courts confront new issues, they consider
the rules and rationales that they announced in prior cases in deciding how the
rules should apply. Sometimes, the courts decide that new rules are appropriate.
COMMON-LAW REGULATION
This course deals with legislation and regulation, which suggests that it is about
a different kind of regulation—rules prescribed by legislatures, rather than
rules announced by courts. However, for centuries, the courts have applied
common-law tort and contract principles—and, for that matter, common-law
property principles—to regulate private behavior and choices. This raises
a question: What is distinctive about legislation as a form of regulation?
One way to consider that question is by looking at the virtues and drawbacks
of the common law as a means of achieving regulatory goals. Consider the
famous (and troubling) case of the Ford Pinto, an affordable subcompact car
that was popular in the 1970s.
In 1972, Lilly Gray’s Pinto stalled as she tried to merge onto a California
freeway. Another car traveling about 30 miles per hour rear-ended her.
The Pinto’s gas tank ruptured,
releasing vapors that spread
through the car.

A spark ignited the mixture, and the Pinto exploded in a ball of fire. Gray
died a few hours later. Her passenger, a 13-year-old boy named Richard
Grimshaw, suffered disfiguring burns and had to undergo facial reconstruction
surgery.
Grimshaw and Gray’s estate filed suit against Ford, claiming that the Pinto’s
gas tank was defectively designed and that the car was unreasonably dangerous.
There was evidence that Ford had cut some corners in production; the
company had designed the Pinto and brought it to market in just 25 months,
considerably faster than the typical new car model, because the company
was worried about losing market share to low-cost options manufactured by
companies in Germany and Japan.
Internal company documents showed that Ford had performed more than
40 crash tests with the Pinto before it became available on the market, and
that the Pinto’s fuel tank had ruptured in every single test conducted at speeds
over 25 miles per hour. Ford engineers considered several low-cost design
fixes, but none made it into the car because the executive vice president
of Ford insisted that the car weigh less than 2,000 pounds and cost less
than $2,000.
After hearing this evidence, the jury in the suit awarded $2.5 million to
the Grimshaws and $500,000 to Gray’s estate to compensate them for their
injuries. In addition, the jury awarded punitive damages of $125 million,
concluding that Ford had knowingly endangered the lives of thousands of
Pinto owners. (An appellate court later reduced the punitive damages award
to $3.5 million.)
This example suggests several things about the common-law tort system
as a means of regulation in this case of potentially dangerous products. On
the one hand, by imposing significant liability on Ford, the system created
a powerful incentive for car manufacturers—and manufacturers of other
dangerous products—to take steps to make their products safe for their
intended uses.

On the other hand, the tort system’s focus on liability after the fact meant
that the Pinto was available on the market notwithstanding its dangers. That
means that millions of people unknowingly faced a meaningful risk of serious
injury or death—and in fact, at least dozens were killed and hundreds more
were injured as a direct consequence of the Pinto’s defects.
Ford was not compelled by government regulation to install the cheap parts
that would have avoided death and injury to scores of Pinto drivers (not to
mention other drivers on the road unlucky enough to get into accidents with
Pintos). Instead, it sold its dangerous cars and then paid damages to people
who were injured by its design choices. Indeed, it simply viewed tort damages
as a cost of doing business.
ASBESTOS
Despite its drawbacks, there are certain important advantages of regulation

through the common-law system. The common-law system is flexible: Judges
can modify the rules to respond to changing circumstances and to advance
the goals of compensation, deterrence, and efficiency.
In other words, both the tort system and legislative action have virtues as forms
of regulation. So when should legislatures act to displace the common-law
tort system in pursuit of regulatory goals?
This question can be examined through

the lens of another thorny case about
product liability. In the 19th century,
scientists discovered the insulating and
fireproofing properties of asbestos.
By the middle of the 20th century, the
product was widely used in manufacturing
and building construction, not only as
insulation, but also to make cement
stronger without adding much weight.
However, in the 1970s, it had become clear
that exposure to asbestos fibers causes
lung cancer and other fatal diseases.

In one sense, these circumstances, although obviously tragic for thousands
of families, seem like the bread and butter of tort litigation. Under modern
product-liability rules, the manufacturer of an unreasonably dangerous product
is liable to persons injured by the product, even if the manufacturer exercised
all due care in the preparation and sale of the product.
However, there were several serious obstacles to providing a remedy through

the tort system to people who were injured by exposure to asbestos. First,
the statute of limitations was a problem. Under traditional tort doctrine,
an injured person has to file suit within some fixed period of time after
suffering the injury; in most states, the statute of limitations for a tort claim
is three years.
But asbestos exposure is not like a car accident; there is not one definable
moment when the exposure has caused an injury, and many people who were
exposed didn’t even know it. In fact, there often was a long time between
exposure and the manifestation of the injury.

Second, causation was an issue. A plaintiff who sues in tort must prove by
a preponderance of the evidence not only that the product was unreasonably
dangerous, but also that the defendant’s
product caused the injury. This is easier said
than done in the case of asbestos exposure.
Some conditions caused by asbestos exposure,
such as lung cancer, can also be caused by
other products, such as cigarettes.
Third, the sheer number of potential plaintiffs threatened to overwhelm the

court system. Fourth, it wasn’t clear that all of the hundreds of thousands of
potential claimants could receive compensation, even if they could overcome
these other problems.
Under traditional tort doctrine, the amount of damages that a plaintiff receives
is the amount necessary to make the plaintiff whole. Given the severity of
the diseases caused by asbestos exposure, successful plaintiffs likely would
recover substantial amounts of money. There was a risk that the first wave
of plaintiffs who recovered damages from asbestos manufacturers would
bankrupt those companies, leaving nothing for the thousands of others who
developed asbestos-related diseases.
THE COURTS’ RESPONSE
How should courts deal with these obstacles? One of the virtues of the
common-law approach to regulation is its flexibility. For example, courts
modified the typical approach to statutes of limitations by creating a discovery
rule, under which the statute of limitations would not start running until the
injured person discovered, or should have discovered, that the person had
been injured.
In addition, some courts considered addressing the problem of causation

by assigning liability to each defendant based on the share of the asbestos
market that it had supplied, unless a defendant could prove that it had not
supplied the asbestos to which the plaintiff was exposed. Most courts rejected
this approach.

Some courts tried to deal with the problem of the sheer volume of asbestos-
related claims by certifying class actions. A class action is a suit by one or
a few plaintiffs on behalf of a class of similarly situated persons. A class action
binds all of the members of the class, even if they do not actively participate
in the litigation. However, courts generally do not permit class actions to
proceed when the individualized questions, such as exposure, causation, and
the extent of the injuries, predominate over the common questions, such as
whether asbestos is an unreasonably dangerous product.
Finally, there was no perfect judicial solution to the problem of asbestos

manufacturers going bankrupt. Plaintiffs began to file suits against other
plausible defendants, such as the construction firms that installed asbestos,
or the car manufacturers that used asbestos in their braking systems, or the
companies for which the plaintiffs worked and where they were exposed to
asbestos. That just brought the courts back to the same problems.
LEGISLATION
If the common-law tort system as applied by courts is not the right vehicle
for solving the problem, then what is? That brings this lecture to the topic
of legislation. Legislatures are not limited to a set of common-law rules that
govern unless there is some compelling reason to depart from them. They can
craft—subject only to constitutional limitations—whatever regulatory response
best accommodates the competing interests implicated by the problem that
they are addressing.
Still, legislation is imperfect. Even legislative solutions inevitably leave some

important questions unresolved. Even the most conscientious legislature
cannot anticipate every question that will arise, and judges eventually will
have to answer them.
Second, consider the actual outcome of our legal system’s decades-long

odyssey with asbestos. While courts grappled with around 1,000,000
claims arising from asbestos exposure, Congress repeatedly considered
a legislative solution.

Bills introduced in Congress would have done something similar to the
measures listed above: created a fund of over $100 billion, financed by
contributions from manufacturers and insurance companies, to provide
compensation on a no-fault basis to persons who are suffering from illnesses
caused by exposure to asbestos.
None of the bills ever became law. Many insurance companies lobbied against
the bill because, in light of the bankruptcy of the manufacturers of asbestos,
they would be on the hook for a large share of the contributions to the fund.
Some lawyers for plaintiffs objected that the limits that the bill would impose
on recovery were unfair. The problem of asbestos proved to be too much for
Congress, too.
Suggested Reading
<< Landes and Posner, The Economic Structure of
Tort Law.
<< Sunstein, Valuing Life.
ÞÞ How does regulation by statute differ from regulation by

the common law?
ÞÞ Why might we choose to regulate by enacting a statute

rather than by leaving the issue to resolution by courts
applying tort or contract law?

Lecture 3
LEGISLATION AND THE

ADMINISTRATIVE STATE
T
he basic system of the American government involves three branches
of government: legislative, executive, and judicial. Put simply, the
legislature enacts the law, the executive enforces the law, and the
judiciary interprets the law, all as part of a system to promote the public
interest. In the real world, matters are more complicated. An understanding
of the legislative process and our regulatory system is crucial to understanding
how we resolve disputes over the meaning of the law and the scope of
government regulation.
THE LEGISLATIVE PROCESS
One way to understand the process for enacting law in the United States is to
compare it to the process in most other Western democracies. Many modern
democracies use a parliamentary system. In those systems, the executive—often
the prime minister—is from the same party that controls the legislature, and
she holds office precisely because her party won control of the legislature. That
means that when the majority in the legislature supports the enactment of a law,
the executive can be expected to support it, too. In addition, parliamentary
systems often have a unicameral—or one-house—legislature, and there
typically are very few opportunities for members of parliament who oppose
legislative proposals to derail those proposals.
The system in the United States is different. The president is elected separately,
which means that the president can be from a different political party than
the majority in Congress. The president’s assent is necessary for a bill to
become law, unless both houses of Congress can muster a super-majority
vote to override the president’s veto. Additionally, America has a bicameral
legislature, and a bill can become law only if both houses of Congress
approve it.
Each house follows rules that allow multiple opportunities for opponents to
kill proposed legislation. In other words, it is much more difficult to pass
legislation in the United States—at least at the federal level—than it is in
other countries.
THE CIVIL RIGHTS ACT
To see how this is works in practice, this lecture now turns to the history
of the Civil Rights Act of 1964, the most significant piece of civil rights
legislation in American history. Almost 10 years after the Supreme Court
held in Brown v. Board of Education that official segregation violates the
Constitution, the promise of that decision had not yet been realized because
of official and private resistance to integration efforts.

In 1963, a quarter of a million
people, including Martin
Luther King Jr., marched on
Washington DC to demand
legislation to protect civil
rights. President Kennedy
proposed the enactment of
such legislation.
The Democrats controlled both houses of Congress and the White House,
but the party was divided. A bloc of northern Democrats strongly supported
legislation to protect civil rights, particularly of African Americans, who had
been the victims of discrimination for well over a century. These legislators
wanted to prevent discrimination in many settings, including restaurants,
hotels, and the workplace.
A smaller bloc of southern Democrats opposed any civil rights legislation.

They had twice tried to obstruct the enactment of civil rights legislation in
the prior decade. Many Republicans strongly supported the cause of civil
rights, but they were also concerned that extending civil rights protections
to the workplace would impose significant costs—litigation related and
otherwise—on businesses.
Because of Congress’s unique rules about enacting legislation, Democratic

supporters of the bill needed some Republican votes to overcome the opposition
of a relatively small but determined minority of legislators. The law that
Congress eventually enacted banned discrimination in public accommodations,
by institutions receiving federal funds, and in the workplace. It also created
the Equal Employment Opportunity Commission (EEOC).
Lecture 3–Legislation and the Administrative State 21

THE WORKPLACE PROTECTIONS
The workplace protections were contained in Title VII of the statute. Title
VII prohibits workplace discrimination on the basis of race, sex, religion, or
national origin.
There was considerable debate about the content and scope of these provisions.
But there was also lots of procedural maneuvering. The process of crafting
legislation does not entail only debate and compromise over the substance of
the law; it also involves navigation of the complex procedural rules that the
two houses of Congress have adopted to govern the legislative process.
Several measures can obstruct the passage of a bill. To take just a handful
of examples, a bill can be killed by inaction by the committee charged with
formulating its details, during consideration on the floor, by a filibuster or
a poison-pill amendment designed to weaken its ultimate support, by the
failure of one house to act on the bill, in a conference committee between
the two houses, or by presidential veto. For example, southern senators
used the filibuster to slow down—and, they hoped, prevent passage of—the
Civil Rights Act.
Southern Democrats filibustered the civil rights bill in the Senate. To invoke
cloture and break the filibuster, supporters of the bill needed the votes of 67
Senators. The Democrats held 65 seats in the Senate in 1964, but because of
the ideological split between northern and southern Democrats, those who
favored the legislation needed Republican support to break the southern
Democrats’ filibuster.
That gave Republican senators considerable leverage. Many Republican senators

agreed with northern Democrats about the need for civil rights protections,
but disagreed with them about the costs of regulating business. Republicans
used their leverage to obtain protections for businesses.
The sponsors of the bill, for example, agreed to include a provision limiting
the power of the EEOC to issue rules governing the workplace, a provision
expanding the defenses available to employers facing claims of discrimination,
and a provision making clear that the law does not require affirmative action.

With those changes, many Republican senators agreed to vote to end the
filibuster and then ultimately to vote in support of the bill. Although they
held only 35 seats in the Senate, the Republicans got the protections they
wanted for business.
AFFIRMATIVE ACTION
This messy process of procedural maneuvering and compromise is important

when it comes to the eventual inquiry about what a statute actually means.
Consider the question of race-conscious affirmative action in the workplace.
The centerpiece of Title VII of the Civil Rights Act states that it shall be
an unlawful employment practice for an employer to refuse to hire any
individual, or otherwise discriminate against any individual, because of such
individual’s race.
Another provision states that nothing in Title VII “shall be interpreted to

require any employer to grant preferential treatment to any individual or
to any group” because of race “on account of an imbalance” in comparison
to “the available work force.”

In the years after the enactment of the Civil Rights Act, many employers
adopted voluntary affirmative action policies. Under those policies, employers
gave preference in hiring and promotion to African American applicants and
other applicants from historically disadvantaged minority groups. White
applicants who were not hired or promoted challenged some of those policies,
claiming that they violated Title VII.
The white applicants pointed to the plain language of Section 703(a), which
says that the refusal to hire a person because of his race, or otherwise
discriminate because of race, is unlawful. Notably, another section, 703(j),
is the provision that makes clear that employers are not required to adopt
affirmative action programs.
When the Supreme Court addressed this question in a 1979 case called
United Steelworkers of America v. Weber, it concluded that Title VII does
not prohibit private employers from using voluntary race-conscious
affirmative action in hiring. The court read the statute’s general prohibition
on discrimination on the basis of race in light of the overall purpose of the act,
which was to expand employment opportunities for historically disadvantaged
minority groups.

The focused on the specific legislative bargain that produced Section 703(j),
the provision that makes clear that it does not require affirmative action. The
court reasoned first that, if Congress had meant to prohibit affirmative action
plans, it would not simply have said that the act does not require such plans.
It instead would have said that it should not be read to “require or permit”
such policies.
The court also noted that senators who were skeptical of the original bill
worried that the Equal Employment Opportunity Commission (EEOC) would
interpret the law to require employers to grant preferential treatment to create
racially balanced workplaces. The Court cited statements from Republican
senators who argued that the law should prevent undue federal government
“interference with private businesses because of [the EEOC’s] ideas about”
racial balance.
In the court’s view, a prohibition against all voluntary affirmative action efforts
would undermine this goal. That is because it would “augment the powers of
the Federal Government and diminish traditional management prerogatives
while at the same time impeding attainment” of the law’s ultimate goals of
ending discrimination.
Justice Rehnquist dissented. He argued that the

plain language of the law and its legislative
history make clear that Congress meant
to prohibit race-conscious affirmative
action policies.
The takeaway point here is that the bargains

that competing factions in Congress made to
overcome procedural barriers can provide strong
evidence of statutory meaning. If that is true, then
we can expect courts to look at more than simply the
text of the law when they interpret statutes.

ADMINISTRATIVE AGENCIES
Congress often gives enforcement power over a particular statute to an

administrative agency. When courts interpret statutes, in other words, they
are often not writing on a clean slate: The agency charged with implementing
the statute might have already interpreted the statute in its effort to carry out
the statute’s objectives.
This raises an important question: What if an agency with broad authority to

issue rules to implement a vaguely worded statute has already resolved some
question about what the statute means, and a person subject to enforcement
by the agency challenges the agency’s interpretation of the statute?
Consider again Title VII of the Civil Rights Act of 1964. The statute also
prohibits workplace discrimination on the basis of sex. The original version
of the bill considered in the House of Representatives did not prohibit
discrimination on the basis of sex. Advocates of women’s rights lobbied
Congress to expand the bill’s protections, but the bill’s sponsors worried
that adding sex discrimination would erode the bill’s support. Some of the
bill’s sponsors also thought that sex discrimination was not a problem that
required a national solution.
However, when the original bill was debated on the House floor, Congressman
Howard Smith of Virginia, a prominent opponent of the bill, proposed an
amendment that would add sex to the list of prohibited grounds for workplace
discrimination. It is clear from Smith’s record that he was a supporter of
equal rights for women, but he also wanted the Civil Rights Act to fail. He
thought that by adding a provision banning sex discrimination, the bill was
less likely to pass.
A coalition of southern Democrats, who opposed the bill, and Republicans

who supported equal rights for women supported the amendment, and it
passed. Contrary to Smith’s hopes, the Civil Rights Act as amended later
passed as well.

THE EEOC AND THE WORKPLACE
Additionally, in the act, Congress gave the EEOC power to

help resolve disputes with employees and authorized it to bring
lawsuits on behalf of employees to enforce the employment
discrimination provisions of the law. Not long after the
enactment of the law, the EEOC issued a report concluding
that the law’s prohibition on sex discrimination in the
workplace also requires employers to provide reasonable
job protection to female employees who become pregnant.
At the time, many employers had policies that burdened

employees who got pregnant. The EEOC reasoned that,
because only female employees can get pregnant, these
policies amounted to impermissible discrimination on
the basis of sex.
If a pregnant employee sued her employer, arguing that one

of these policies constituted unlawful sex discrimination
under Title VII, what role should the EEOC’s view of the
statute have played in the court’s interpretation of the statute?
The EEOC was filled with government employees who gave
considerable thought to questions of workplace fairness,
and they presumably had conducted research about the
practical effect of pregnancy-related policies on female
employees.
Should the court, which had to decide whether the policy was “discrimination
on the basis of sex,” defer to the agency’s view about the statute’s meaning? Or
should the court simply engage in the same process identified earlier, simply
asking what Congress meant when it banned sex discrimination?
It turns out that the EEOC does not have power to issue binding rules to
implement Title VII. As a consequence, perhaps Congress did not want the
agency to have any unique power to issue conclusive interpretations of the
Civil Rights Act.

For another example: The Federal Trade Commission (FTC) does have power
to issue rules with the force of law. Imagine that the FTC issues a rule that
makes it an unfair trade practice for a funeral home to require a family seeking
cremation services to purchase a casket in which cremation will occur.
The FTC then brings an action against a funeral home that has violated the
rule to force it to comply and to impose fines for the violation of the rule.
The funeral home defends by arguing that the Federal Trade Commission
Act, one of the statutes that the FTC enforces, is not properly read to prohibit
this practice.
Should the commission’s interpretation of the statute bind the court? Even if
not, should the court defer to the Commission’s view, accepting it as long as
it’s a reasonable interpretation of the federal statute?
To resolve those questions, it is necessary to learn more about statutory

interpretation by courts, the delegation of power to agencies, and the limits on
Congress’s power to create agencies that enjoy some level of independence from
presidential control. Those topics are considered in the lectures that follow.
Suggested Reading
<< Oleszek, Congressional Procedures and the Policy
Process.
<< Whalen and Whalen, The Longest Debate.
ÞÞ Are the obstacles to enacting federal legislation a feature

of our system or a bug?
ÞÞ How should the complexities of our system for enacting

federal legislation affect courts’ interpretation of the
statutes that make it all the way through the system?

Lecture 4
TOUCHSTONES OF
STATUTORY
INTERPRETATION
A
n essential skill for understanding legislation and regulation is
statutory interpretation. Statutes are written laws enacted by
a legislature. That means that at the heart of statutory interpretation
is the reading of text.
At the heart of interpretation are four touchstones methods that people apply:
zz Sometimes, we simply follow the plain

meaning of the words.
zz Sometimes, we need to consider context

to figure out what the text means.
zz Other times, we have to think about the

intent or purpose of the words.
zz Still other times, it seems

like the only way faithfully
to follow instructions is to
depart from them.
To delve into interpretation, this

lecture focuses on a notable Supreme
Court case.
CHURCH OF THE HOLY TRINITY V. UNITED STATES
One of the most famous statutory-interpretation cases of all time was Church
of the Holy Trinity v. United States, which the Supreme Court decided in 1892.
The church had entered a contract with Walpole Warren, a British minister,
to become the pastor of its congregation. Warren came to the United States
and began to work for the church. The United States then brought an action
to impose a penalty on the church for violating the Alien Contract Labor Law
of 1885. A lower court ruled against the church, invalidating the contract, and
the church appealed the decision to the US Supreme Court.
Sometimes when Congress enacts a law, it includes a title (or description) of the
law. The title provision of this law called it “An Act to prohibit the importation
and migration of foreigners and aliens under contract or agreement to perform
labor in the United States.”
Section 1 of the law stated that “it shall be unlawful for any person or
corporation to prepay the transportation, or in any way assist or encourage
the importation or migration of any alien into the United States, under
contract or agreement to perform labor or service of any kind in the United
States.” Section 5 of the statute said, “the provisions of this act shall not apply
to professional actors, artists, lecturers, or singers, nor to persons employed
strictly as personal or domestic servants.”

The question for the court was whether the church’s contract with Warren
violated the statute. The court began its opinion by declaring, “It must be
conceded” that the church’s act “is within the letter” of Section 1 of the statute,
“for the relation of rector to his church is one of service, and implies labor on
one side with compensation on the other.”
The court noted that the statute prohibits contracts to bring over aliens
who will perform labor or “service of any kind.” The court acknowledged
that Section 5 of the law contains “specific exceptions,” which “strengthens
the idea that every other kind of labor and service was intended to be reached
by the first section.” In other words, the Court began by noting that the plain
meaning of the words of the statute embraced the church’s contract with
the pastor.
However, the court nevertheless stated, “we cannot think Congress intended
to denounce with penalties a transaction like that in the present case.” The
court cited what it called a “familiar rule” that “a thing may be within the
letter of the statute and yet not within the statute, because not within its spirit,
nor within the intention of its makers.”
THE COURT’S REASONING, PART 1
To support its claim that punishing the church for its contract with the pastor
would be contrary to the spirit of the statute or Congress’s intent, the court
cited three pieces of evidence. First, the Court pointed to the title of the act,
which referred only to contracts to “perform labor” (but not to contracts
for “service”).
In the court’s view, the phrase “perform labor” implied “only the work of
the manual laborer, as distinguished from that of the professional man” or
“any class whose toil is that of the brain.” The court said that the “common
understanding of the terms ‘labor’ and ‘laborers’ does not include preaching
and preachers,” and so the language of title provision suggested that ministers
were excluded from the ban.
Lecture 4–Touchstones of Statutory Interpretation 31

Second, the court stated, “another guide
to the meaning of a statute is found in
the evil which it is designed to remedy.”
The court declared that it was “common
knowledge” that the impetus for the law
was the practice of large companies to
“contract with their agents abroad for
the shipment of great numbers” of low-
skilled laborers; the companies would
pay for the workers’ transportation to
the United States, and in return the
workers would agree to work for very
low wages.
According to the court, the effect of

these practices was to depress wages
for “other laborers engaged in like
occupations.” The court also noted that
Congress enacted the law after Congress
received petitions and heard testimony
about the threat that cheap, unskilled
labor posed to the labor market. The court stated that “it was never suggested”
that the country had “a surplus of brain toilers and, least of all, that the market
for the services of Christian ministers was depressed by foreign competition.”
Third, the court relied on the report of the Senate Committee on Education
and Labor recommending the passage of the bill. The report stated that, if not
for the press of time, the committee would have recommended an amendment
to the bill substituting the phrase “manual labor” or “manual service” for the
phrase “labor or service.”
The committee did not propose the amendment, according to the report, “in
the hope that” the bill would “become a law before the adjournment” of the
legislative session. In addition, the report stated the committee’s expectation
that the phrase “labor or service,” which it left unchanged, “will be construed
as including only those whose labor or service is manual in character.”

Based on this evidence—the title of the act, the evil that Congress intended to
remedy, and the act’s legislative history—the court concluded that “the intent
of Congress was simply to stay the influx of this cheap, unskilled labor”—that
is, not skilled professionals like Warren.
THE COURT’S REASONING, PART 2
Just in case there was any doubt, the court also asserted that “no purpose
of action against religion” should “be imputed to any legislation” because
Americans are “a religious people.” The court asked the reader to imagine
what would have happened if a member of Congress had proposed a bill that
prohibited the Catholic Church from contracting to bring a cardinal to the
United States or a Jewish synagogue from seeking to bring an eminent rabbi to
the country. In the Court’s view, such a bill would not have received “a minute
of approving thought or a single vote.” As a consequence, it made no sense to
read the statute to permit such a result.
In construing this relatively brief statute, the court considered several

of the touchstones for statutory interpretation mentioned earlier.
The court started with the plain meaning of the words in the statute,
seeming to go out its way to establish that the hiring of Warren fell within
the prohibition in the act.
The court conceded that Section 5, which listed exempted professions, supports
the conclusion the ministers are covered by the plain meaning of Section 1
because Section 5 does not expressly exclude ministers from coverage. This
raises a question: Is it obvious that the omission of the word ministers from
Section 5 means that they are covered by the language in Section 1?
The court also touched on the act’s purpose. In focusing on the evil that the
law was designed to remedy, the court essentially reasoned that the statute’s
purpose was to exclude only a certain type of employee. Its provisions
should be read in light of that purpose, notwithstanding the law’s plain
language. Notably, this case was the first time that the Supreme Court relied
on legislative history to reach a conclusion at odds with the plain meaning
of a statute.
Lecture 4–Touchstones of Statutory Interpretation 33

Finally, the court departed from the text by relying on a background norm
against interference with religion. The court seemed to suggest that the norm
mattered because, in cases of doubt about statutory meaning, such norms
should tip the balance.
Suggested Reading
<< Chomsky, “Unlocking the Mysteries of Holy
Trinity.”
<< Dickerson, The Interpretation and Application

of Statutes.
<< Eskridge, Dynamic Statutory Interpretation.
<< ———, “Public Values in Statutory Interpretation.”
<< Popkin, Statutes in Court.
<< Vermeule, “Legislative History and the Limits of

Judicial Competence.”
ÞÞ Do we use the same tools when we interpret a statute that
we use when we interpret any other text?
ÞÞ When we interpret texts, are we always using the same

method—for example, seeking the plain meaning of the
words—or do we instead vary our approach depending
on the circumstances?

Lecture 5
THE LETTER VERSUS

THE SPIRIT OF THE LAW
T
he previous lecture introduced the touchstones of interpretation—
the various indications of meaning that we consider when we try to
interpret a set of instructions. This lecture turns to specifically examine
purpose as a touchstone for meaning by looking at the opinions surrounding
Riggs v. Palmer.
RIGGS V. PALMER
The famous case of Riggs v. Palmer was decided in 1889. New York’s statute of
wills governed the disposition of property after a person’s death. The statute
specified the rules for making an enforceable will, such as the requirement that
it be in writing and that witnesses attest that the drafter signed it. The statute
also said that anyone named in a will as a beneficiary would inherit “except
in cases of fraud, duress, or incapacity” at the time the will was made.
In other words, a bequest in a will wouldn’t be enforceable if the beneficiary had
duped or forced the person into making the bequest. Additionally, a bequest
in a will would be unenforceable if the person who made the will was not of
sound mind. Otherwise, the statute stated quite clearly that the terms of the
will would govern.
Francis Palmer wrote a will leaving some specific property

to his two daughters and the remainder of his estate to his
16-year-old grandson, Elmer. Elmer learned that he stood
to inherit, and he could not wait to get his hands on the
property. He was also concerned that his grandfather
might change his will. Elmer killed his grandfather by
putting poison in his food.
Elmer was arrested, prosecuted, and then convicted of

second-degree murder. However, when the will went to
probate—the judicial process for sorting out who gets
what after someone dies—Elmer’s lawyer appeared and
asserted a claim to most of the estate. He pointed out that
Francis was of sound mind and that there had been no fraud or duress in the
creation of the will. Under the plain terms of the will, Elmer was entitled to
inherit most of Francis’s estate.
Elmer’s aunts—Francis’s two daughters who also were beneficiaries of the

estate—objected. They sought to annul the will, arguing that a murderer
should not be entitled to inherit from his victim. To even suggest that he
should, they said, is absurd.
Notice the question of statutory interpretation that the case raised for
the court. The statute itself was quite clear. It said that a person named
as a beneficiary in an otherwise valid will shall inherit, except in cases of
fraud, duress, or incapacity. No one contended in the case that any of those
exceptions applied.

RIGGS V. PALMER IN
THE NEW YORK COURT OF APPEALS
The New York Court of Appeals—New York’s highest court—concluded

that Elmer was not entitled to inherit under the statute of wills. The
court acknowledged that the letter of the law seemed to require that Elmer
inherit most of Francis’s estate. However, the court declared that “a thing
which is within the intention of the makers of a statute is as much within
the statute as if it were within the letter; and a thing which is within the
letter of the statute is not within the statute unless it be within the intention
of the makers.”
The court explained that the purpose of the statute of wills was to enable people
with wills “to dispose of their estates to the objects of their bounty at death,
and to carry into effect their final wishes legally expressed.” In other words,
the legislature wanted to be sure that the wishes of people who follow the rules
in disposing of their estates would be honored upon their deaths.
The court said that it was clearly the legislature’s intention that, as a general
matter, people named in a will should get the property left to them by the
will. But, the court said, “it never could have been their intention that
a [beneficiary] who murdered the testator to make the will operative should
have any benefit under it.”
The court explained that it is not possible for a legislature to anticipate every
case that might arise and deal with it expressly in the statute. When faced with
a question like the one in Palmer’s case, the court said that we should imagine
that the lawmakers were present and that we could ask them: Did you mean
to let Elmer inherit under these circumstances?
The court in Riggs enforced the spirit of the law, not the letter of the
law. Because the literal text of the law required an unreasonable result in
tension with the law’s spirit, the court declined to give effect to the law’s
plain terms.
Lecture 5–The Letter versus the Spirit of the Law 37

OTHER OPINIONS
The New York court concluded that Elmer could not inherit because such
a result would be contrary to the spirit of the law. However, the New York
court was not the first court to confront this problem; sadly, the prospect
of inheritance had prompted people in other states to commit similar acts.
Virtually all of the other courts that had confronted this issue had concluded
that the plain terms of the will, and the plain terms of the statute of wills,
had to be given effect, notwithstanding the wrongdoing of the heir.
Why would they reach that conclusion? Consider the opinion of Justice Gray,
who dissented in Riggs v. Palmer. Though he, too, was troubled by Elmer’s
actions, he made two basic arguments about why the court should follow
the letter, rather than the spirit, of the law.
First, he asserted that the judge’s role is to follow the plain words of the law,
not to change the law when the judges think that the legislature’s judgments
seem misguided. Second, he asserted that there is virtue in preserving the
clarity of bright-line rules, even though in some cases they will produce
seemingly unfair results.
Justice Gray’s first objection had to do with the proper judicial role in
interpreting statutes: Legislatures make the law, the executive branch enforces
the law, and the courts interpret the law. If courts could ignore what the
legislature has said, then they would not be interpreting the law; they would
be making it.
Justice Gray’s second (and related) objection to the court’s approach had
to do with the potential virtue of clear rules and the problem of judicially
created exceptions to clear rules. Clear rules about who can inherit under
a will, and the circumstances under which wills are valid and enforceable, are
important because they let us plan with certainty, confident that our wishes
will be honored.

Justice Gray’s basic point was that the statute of wills was designed to maximize
the chance that we would give effect to the wishes of the person who wrote
the will. The court did not seem to disagree about this; it shared Justice Gray’s
assumption about the point of the statute.
The court, in fact, was trying to give effect to Francis Palmer’s wishes; after
all, Francis surely would not have wanted his estate to pass to his grandson
if he had known that his grandson would kill him for the money. In light
of this, shouldn’t the court be permitted to craft an exception to the law to
effectuate its basic purpose, in spite of Justice Gray’s concern?
Justice Gray believed that, once we authorize judges to depart from the plain
meaning of enacted law to reach just results, there is no logical stopping point.
It begs the question: When judges do so, are they enforcing the legislature’s
will or their own views about what is just and appropriate?
CONCLUSION
The debate between the majority and Justice Gray in Riggs was not over the
object of statutory interpretation. They agreed that the goal of statutory
interpretation is to act as the faithful agent of the legislature.

The debate instead was over the method: How should a court determine
what the legislature’s will actually was? The majority thought that the best
method was to discern the legislature’s purpose and then to read the statute
so as to carry out that purpose. Justice Gray thought that the best approach
was to treat the plain words of the statute as the best and only indication of
the legislature’s will.
This debate continues today. Some scholars and judges think that the best way
to implement the legislature’s wishes is for courts to discern the legislature’s
purpose and then to read the statute in light of that purpose. Others think
that confining the inquiry to the statute’s plain meaning is the only way to
ensure that judges do not stray across the line between interpretation and
policy making.
In practice, these two approaches tend to overlap. Judges who seek legislative
purpose look, first and foremost, to the statutory text to discern the legislature’s
purpose. Judges who rely on the letter of the law often consider the background
against which the statute was enacted to determine the plain meaning of the
words of the statute.
Additionally, legislative purpose involves several distinct (albeit related)

concepts. Sometimes courts ask what the legislature’s intent was with respect
to a particular question. Courts might try to determine the legislature’s intent
by considering specific evidence about what members of the legislature thought
about the question. Other times, courts seek the legislature’s broader purpose,
asking why the legislature enacted the law or what problem the legislature
sought to address, and then seek to answer the specific question in light of
that purpose.
These two inquiries also often overlap in practice. In some cases, the best
evidence of the broader statutory purpose will be what individual members
of the legislature said about the types of cases that are likely to arise. In other
cases, a court will conclude that the legislature intended to resolve a specific
issue in part because of more general indications of legislative purpose.
In Riggs, for example, the court talked about the legislature’s “intentions,”
but it really sought to determine the legislature’s broader purpose—or, as the
court described it, its “spirit.”

Suggested Reading
<< Dworkin, Law’s Empire.
<< Hart and Sacks, The Legal Process.
<< Radin, “A Short Way with Statutes.”
<< Scalia and Garner, Reading Law.
ÞÞ How would you go about determining the purpose of the

legislature that enacted a statute?
ÞÞ Is it possible for a judge to determine legislative purpose

without substituting her own personal policy preferences
for those expressed in the plain text of the statute?

Lecture 6
WHEN IS STATUTORY
MEANING PLAIN?
T
his lecture considers just how robust our commitment to the plain
meaning of statutes should be. Should departures from the letter of the
law to enforce the spirit of the law be the rare exceptions, rather than the
rule? To delve into that question, this lecture examines the cases of Tennessee
Valley Authority v. Hill and West Virginia University Hospitals, Inc. v. Casey.
BACKGROUND ON
TENNESSEE VALLEY AUTHORITY V. HILL
The Endangered Species Act of 1973 states that all federal agencies shall “insure
that actions authorized, funded, or carried out by them do not jeopardize the
continued existence of such endangered species” or “result in the destruction
or modification of habitat of such species.”
Six years before Congress enacted that law, the Tennessee Valley Authority,
a federal agency, began construction on the Tellico Dam and Reservoir project.
The agency set out to build the dam on the Little Tennessee River, not far
from where the river’s waters meet the Big Tennessee River. Congress first
appropriated money for the project in 1967, and did so each year through
1976. By 1975, construction on the dam was almost complete, but it did
not open.
In 1973, a scientist at the University of Tennessee had discovered a previously

unknown species of perch called the snail darter, a three-inch, tan-colored
fish. Scientists, and eventually the secretary of the interior, concluded that
there were only about 10,000 snail darters left.
In 1975, around the time that the Tennessee Valley Authority was nearing
completion of the project, the secretary formally designated the snail darter
as an endangered species. The secretary determined that the snail darter lived
only in the portion of the Little Tennessee River that would be completely
flooded by the Tellico Dam and Reservoir project, and that the flooding would
result in the total destruction of the snail darter’s habitat.
A regional association of biologists dedicated to the preservation of local

species filed suit, seeking to prevent the flooding of the snail darter’s habitat
and the opening of the dam. The plaintiffs argued that the completion of
the dam and reservoir would violate the Endangered Species Act by directly
causing the extinction of the snail darter.
Shortly thereafter, the House and Senate Appropriations Committees, which

were considering the budget for the upcoming year, held hearings about the
continued funding of the project. The chairman of the Tennessee Valley
Authority testified that he didn’t believe that the Endangered Species Act
applied to a project that was more than 50 percent complete at the time the law
was enacted and almost 80 percent complete at the time that the snail darter
was designated as an endangered species. Congress then passed a budget bill
appropriating $9 million for continued work on the project.
Lecture 6–When Is Statutory Meaning Plain? 43

IN THE COURTS
Although the trial court thought that the completion of the project would not
violate the Endangered Species Act, the court of appeals disagreed and halted
work on the project. The Supreme Court then decided to review the case. The
question was whether the Endangered Species Act prohibited the Tennessee
Valley Authority from continuing work on the project.
One way to approach this question of statutory interpretation would be to note

that Congress clearly did not want to prohibit the completion of the project,
which began well before the law’s enactment, because Congress continued to
appropriate money for it even after it learned about the snail darter. On this
view, the proper touchstone of interpretation is congressional intent.
Another (albeit related) way to approach the question is to consider the spirit
of the law. On this view, the correct touchstone of interpretation is legislative
purpose. The appropriate question is: What was the broad, animating principle
that motivated Congress and what reading of the statute’s terms best advances
that purpose?
On this approach, it would not matter if Congress had not actually considered
how the Endangered Species Act would apply to the Tellico Dam project; it is
enough to note that Congress’s actions reflect a purpose to balance the interest
in environmental protection with rural economic development. Killing the
almost-completed dam project would surely be inconsistent with that purpose.
That was the dissent’s approach to the question in the case.
THE COURT’S CONCLUSION
The Supreme Court concluded that the answer to the question presented by
the case was compelled by the plain language of the Endangered Species Act.
“One would be hard pressed to find,” the court said, “a statutory provision
whose terms were any plainer than those” in the act. The statute required all
agencies “to insure that actions authorized, funded, or carried out by them do
not jeopardize the continued existence” of any endangered species or “result
in the destruction or modification of habitat of such species.” The court said
that this language “admits of no exception.”

To accept the government’s position, which was that Congress could not
possibly have meant to limit a project that was well under way when it enacted
the law, the court would be forced to ignore “the ordinary meaning of plain
language.” However, the court could not see how the agency could close the
gates of the dam and flood the reservoir—actions that would threaten the
snail darter—without “carrying out” an action that has been “authorized”
and “funded” by a federal agency. The court also could see how that action
would “insure” that the snail darter’s habitat is not disrupted or destroyed.
That, in sum, was the court’s reasoning for concluding that the dam project
had to cease, notwithstanding the tens of millions of taxpayer dollars that
had already been spent on it. Additionally, because two justices dissented,
asserting that Congress could not possibly have intended this result, the court
looked at the legislative history of the Endangered Species Act to confirm its
view of the text.
A predecessor statute had required federal agencies to protect endangered

species, but it qualified the obligation by stating that they should seek to protect
endangered species only “insofar as is practicable and consistent with their
primary purposes.” The original version of the bill that became the Endangered
Species Act of 1973 also contained such language, and the version of the bill
that the Senate first passed also included such language.
But after pressure from environmentalists, the House passed a bill that omitted
any qualification on the government’s obligation to protect endangered species.
The Conference Committee Report, issued after the two houses had reached
agreement on the final bill, explained that the new law “substantially amplifies
the obligation of federal agencies to take steps within their power to carry
out the purposes of this act.” To the justices in the majority, this legislative
history confirmed that Congress meant what it had said.
The court also made clear that, in the end, the judiciary’s responsibility is
simply to enforce the plain meaning of the words in the statute—and that it
only mentioned the legislative history because the dissenting justices focused
heavily on it.

After the court’s decision, Congress amended the Endangered Species Act
to create a procedure for granting exemptions from the law’s requirements
for projects with significant value to the public. Congress then enacted
a law explicitly exempting the Tellico Dam project from the Endangered
Species Act. Congress, in other words, agreed with the justices who dissented
in the case.
IMPACT ON SNAIL DARTERS
The Tellico Dam was completed, and the snail darter’s habitat in the
Little Tennessee River was destroyed. However, a few years later, the
same scientist who discovered the snail
darter found that it actually
lived in four other rivers in
Tennessee. Although it is
still vulnerable, it is no longer
endangered.
WEST VIRGINIA UNIVERSITY HOSPITALS, INC. V. CASEY
The snail darter case raises an important question: Why would a judge rely
on the plain meaning of a statute when there are strong indications that the
legislature would have wanted a different resolution of the issue presented?
Justice Antonin Scalia, who served on the Supreme

Court from 1986 until his death in 2016, was
arguably the court’s most eloquent proponent
of the view that judges should be bound by
the letter, rather than the spirit, of the law.
One example of that is his opinion in a case
called West Virginia University Hospitals,
Inc. v. Casey.

The ordinary rule in litigation in the United States is that each party pays his
or her own attorney’s fees, regardless of who wins. But the cost of litigation
can be a significant deterrent to otherwise meritorious lawsuits. When such
lawsuits not only compensate the plaintiff but also vindicate some public
interest, the government sometimes permits the winning party to force the
losing party to pay its litigation costs.
The federal statute at issue in the West Virginia hospitals case was one such
provision. It said that the winning party in a civil rights suit against the
government was entitled to recover “a reasonable attorney’s fee” from the
losing party. The hospital sued the governor of Pennsylvania, claiming that
the state’s Medicaid reimbursement rates were too low. The hospital won at
trial, and the trial court awarded it attorney’s fees, including over $100,000
to cover the costs of several experts—accountants and doctors—who assisted
the lawyers in the preparation of the suit and testified at trial.
The question for the Supreme Court was whether the statute’s authorization of
“attorney’s fees” for the winning party included fees for non-attorney experts.
The court, in an opinion by Justice Scalia, held that the statute authorized the
award of fees only for attorneys, not for experts.
ARGUMENTS ON THE CASE
Justice Stevens argued in dissent. He thought that Congress’s clear purpose

was to preserve access to the courts and encourage public interest litigation
by fairly compensating successful civil rights plaintiffs for the true costs of
litigation.
Justice Scalia, who wrote the majority opinion, disagreed. He thought that
the words in the statute—“reasonable attorney’s fee”—did not embrace expert
fees. He reasoned that Congress had enacted many statutes specifically
authorizing the shifting of both attorney’s fees and expert fees, and that as
a consequence, the term “attorney’s fees” alone had a more limited meaning
in the law.

Justice Stevens believed, in essence, that
Congress forgot to cover expert fees. His
operating assumption was that Congress
legislates in haste, often with serious time
pressures, with limited foresight, armed only
with the inevitably indeterminate power
of words.
Justice Scalia’s view, in contrast, was based on

the assumption that statutes are the product of the
process of negotiation and compromise in politics. It
is entirely possible that a majority of members of the relevant committees in
the House and Senate wanted to authorize the recovery of all litigation costs
for successful civil rights plaintiffs.
But given the many opportunities for opponents to obstruct legislation, the
proponents of a broad statute might have been forced to settle for something
less to persuade their reluctant colleagues to vote for the bill. The only real
evidence that we have of the actual compromise that they struck is the language
of the law itself.
The justices’ disagreement about the judicial role flowed naturally from their
disagreement about the appropriate way to understand the legislative process.
To Justice Scalia, departure from the plain language of the statute replaces the
actual compromise that the members of Congress reached with some idealized
law that Congress never could or would have passed.
To Justice Stevens, in contrast, the way to be faithful to Congress’s will is to

assume that Congress acted consistently with its stated purpose. On this view,
interpretation by dictionary, rather than by common sense, risks betraying
Congress’s will—and will produce less coherent and fair laws.

Suggested Reading
<< Bean and Rowland, The Evolution of National
Wildlife Law.
<< Easterbrook, “Text, History, and Structure in

Statutory Interpretation.”
<< Scalia and Garner, Reading Law.
<< Siegel, “What Statutory Drafting Errors Teach

Us About Statutory Interpretation.”
ÞÞ Can you think of times when it would make sense to
depart from the letter of the law and instead to enforce
its spirit?
ÞÞ How do we know when the text of a statute is clear, such

that its plain terms must govern?

Lecture 7
SEMANTIC AND
SUBSTANTIVE
INTERPRETIVE RULES
W
hen a legislature enacts a statute, it does so against the background
of certain conventions about how to understand language. We use
many of them in everyday communication. For example, we all
agree that words have distinctive meanings, and we all follow essential
rules of grammar to structure the way we express ideas. However, there
is an additional set of background understandings that courts rely on in
interpreting statutes. They are the canons of construction, and they are the
focus of this lecture.
BACKGROUND ON CANONS
Courts have devised canons—or rules of interpretation—for reading statutes.

The canons fall into one of two general categories: semantic canons, which
are a set of generalizations about how the English language is conventionally
used and understood, and substantive canons, which are presumptions in
favor of a particular set of outcomes.
To understand the difference, consider McBoyle v. United States, a Supreme

Court decision from 1931. The National Motor Vehicle Theft Act stated that
a person who transports or causes to be “transported in interstate commerce
a motor vehicle, knowing the same to have been stolen, shall be punished”
by a fine or imprisonment.
Section 2 of the statute said that, “when used in this Act,”
the term motor vehicle “shall include an automobile,
automobile truck, automobile wagon, motor cycle, or
any other self-propelled vehicle not designed for running
on rails.” McBoyle was convicted of transporting
a stolen airplane from Illinois to Oklahoma. The
question for the court was whether an airplane
counted as a motor vehicle within the
meaning of the statute.
JUSTICE HOLMES’S
CONCLUSION
In his brief opinion for the court, Justice Holmes

concluded that the statute did not make it a crime
to transport a stolen airplane. Holmes reasoned
that “in everyday speech ‘vehicle’ calls up the
picture of a thing moving on land.” This was
a typical argument about textual meaning—that
we should understand words as they are ordinarily
used, even if the dictionary definition is broader.
However, perhaps because the phrase “any self-propelled vehicle not designed
for running on rails” arguably included planes, Justice Holmes relied on two
canons of statutory interpretation to support his conclusion. First, he noted that
the phrase in question appeared at the end of a list of more specific terms. The
specific terms, covering automobiles, automobile trucks, automobile wagons,
and motorcycles—indicated that “a vehicle running on land” is “the theme.”
In other words, Holmes argued that we should read the broad term at the
end of a list in light of the specific items that precede it. This is a classic
semantic canon of interpretation known as ejusdem generis, meaning “of the
same kind.” The idea is that words derive meaning from context, and that
we—and legislators—tend to group together words that share some common
characteristic.
Lecture 7–Semantic and Substantive Interpretive Rules 51

Second, Justice Holmes essentially concluded that
because it was not clear that the term vehicle in the
statute included airplanes, it would not be fair
to punish the defendant under the statute. This
is a classic substantive canon of interpretation
called the rule of lenity. Under the rule of
lenity, ambiguities in criminal statutes should
be resolved in favor of criminal defendants.
SEMANTIC CANONS
Perhaps the most common semantic canons—and the ones that we most take
for granted—are the grammar canons. When a legislature enacts a law, it acts
against the background of grammatical conventions. One example is the rule
of last antecedent. For example, if a law states that commercial licenses “shall
not be required for boats, tractors, and trucks under three tons,” a court would
conclude that the phrase “under three tons” refers only to trucks—the item
that immediately precedes the modifying phrase.
Questions of statutory interpretation can also turn on things as mundane

as punctuation. Consider O’Connor v. Oakhurst Dairy, which involved
Maine’s law about when workers were entitled to overtime pay. The law said
that employees were entitled to be paid one-and-a-half times their regular
hourly rate for overtime work, but then it listed several exemptions from the
rule. One of those exemptions said that the overtime requirements do not
apply to the “canning, processing, preserving, freezing, drying, marketing,
storing, packing for shipment or distribution
of perishable foods.”
Delivery drivers for a dairy producer

sued, claiming that they were
entitled to overtime pay. The
employer argued that its drivers
quite clearly engaged in the
“distribution” of perishable
foods, because their job was to
transport dairy products.

The parties disputed the meaning of the phrase “packing
for shipment or distribution.” Was it one phrase,
identifying one exempt activity, as the drivers argued?
Or was it in fact two distinct exempt activities, with
the first being “packing for shipment” and the second
being “distribution,” as the employer argued?
Everyone agreed that if there had been a comma

between “packing for shipment” and “distribution,”
then the answer would have been clear: They would
have been two distinct activities, the last two on a long
list of activities that are exempt from the overtime law.
The court acknowledged that the statute could fairly be

read either as the drivers contended or as the employer
urged. The court concluded that, because the statute
was ambiguous, it should be construed to “further
the overtime law’s remedial purposes,” which was to
provide fair compensation to employees who work
more than 40 hours per week. Because the legislature
omitted the comma, the drivers won.
THE OXFORD COMMA
O’Connor v. Oakhurst Dairy turned on the absence of

the Oxford comma—that is, a comma before the final
item in a list of three or more items. For example, the absence
of the last comma can cause problems in a phrase like this
one: “I’d like to thank my parents, Mother Teresa and the
pope.” Without the comma, the sentence could be read
to mean that the speaker’s parents are Mother Teresa and
the pope. With the comma, it is clear that the speaker is thanking four
people: the pope, Mother Teresa, and her two parents.

EXPRESSIO UNIUS
There are several other canons of

interpretation that address problems
associated with lists. One governs what
assumptions we should make about items
that do not appear on a list that a legislature
has taken the time to enumerate.
Sometimes the answer seems obvious. For

example, if a traffic sign says that “No trucks
or buses may use this road,” it seems fairly
clear that cars and bicycles are allowed on
the road. The specific exclusion of two types
of vehicles seems to indicate, by negative
implication, that all other kinds of vehicles
are permitted.
In fact, there is a traditional canon of interpretation that applies in those

circumstances. The canon is known as expressio unius, which is short for
a Latin phrase meaning, “the expression of one thing indicates exclusion of the
other.” Courts often follow this canon when they interpret statutes with lists.
NOSCITUR A SOCIIS
Another canon governs what we should do when the meaning of one

of the words in a statutory list is unclear because of its breadth. For
example, a federal statute gave favorable tax treatment to income “resulting
from exploration, discovery, or prospecting.” A camera manufacturer argued
that it should be permitted to take advantage of the statute for income from
the sale of its patented products because the products were the result of
a “discovery.”
The court reasoned that, when viewed in light of the other two terms in the
statute—exploration and prospecting—it was clear that the word discovery
meant only “discovery of mineral resources.” This is an application of the
canon known as noscitur a sociis, which means, “it is known by its associates.”

The idea is that words in a statute can derive meaning from the words that
they are grouped with. Even if one of the words could, in the abstract, have
a broader meaning, we should construe its meaning in light of the words
around it.
SUBSTANTIVE CANONS
Not every canon of interpretation is a device for determining the semantic

meaning of the words in a statute. There is also a set of canons, known as
substantive canons, which put a thumb on the scale for certain policies,
values, or outcomes. These canons do not purport to give us a certain answer
to the question of semantic meaning; instead, they operate like a series of
presumptions about what to do when a statute is not totally clear.
The substantive canon with the deepest pedigree is the rule of lenity, discussed
briefly earlier in the lecture. This canon requires courts to interpret laws
authorizing criminal punishment strictly; if the law does not clearly outlaw
some conduct, then a person cannot be penalized for it, even if conventional
approaches to statutory interpretation would suggest that the semantic meaning
of the law’s terms embraces that conduct.
Another principle of interpretation that functions like a substantive canon

is the so-called rule against absurdity. Like the rule of lenity, the rule against
absurdity is designed to avoid a particular set of results, notwithstanding
the semantic meaning of the words in the statute. Courts usually enforce the
plain meaning of statutes, but courts sometimes refuse to do so when the plain
meaning of the statute would produce absurd results.
Consider United States v. Kirby, which the Supreme Court decided in 1868.
A federal statute made it a crime knowingly and willfully to “obstruct or retard
the passage of the mail.” Kirby was a local sheriff who executed an arrest
warrant for murder against a mail carrier while he was delivering the mail.
Kirby was prosecuted under the statute, but the Supreme Court concluded
that the statute did not apply under these circumstances. The Court declared
that “general terms” in statutes “should be so limited in their application as
not to lead to absurd consequences.”

The challenge is to figure out just when a particular set of results would be
absurd. For example, what if Kirby had arrested the mail carrier not for murder,
but for littering, and there was no risk of flight? Would it be so clear in that
case that he had not violated the statute, which was designed to ensure the
uninterrupted flow of the mail?
The problem is that there is no such thing as absurdity in the abstract; whether
a particular set of results would be absurd depends on our intuitions about
what the legislature had in mind in the first place. For example, in the case
United States v. Locke, at issue was a federal statute that required a person
who wanted to claim mining rights on federal land to file paperwork “prior
to December 31.”
The court concluded that the statute barred any claim filed on the last day of
the year, even though most people assumed that Congress made a mistake in
saying “prior to December 31” rather than “no later than December 31.” This
highlights the fact that it can be difficult to tell absurd results from ones that
simply seem unfair.
Returning to the Kirby case, an interesting fact is that the case arose in the
border state of Kentucky right after the end of the Civil War. The mail carrier
that Sheriff Kirby arrested, a man named Farris, had led Union forces against
Confederate guerillas during the war. The “murders” that Kirby arrested
Farris for were the killings of Confederate soldiers—including Kirby’s
brother-in-law—by Union soldiers under Farris’s command. A grand jury of
Confederate sympathizers handpicked by Kirby issued the arrest warrant, and
when Kirby arrested Farris, he was accompanied by an angry mob.
Does that change your view about whether application of the federal statute
to Kirby was absurd? Alternatively, does it simply tend to underscore that, at
bottom, we are making some kind of value judgment in deciding whether the
plain meaning of a statute produces absurd results? The takeaway point is that
even when we have canons of interpretation to guide us, the task of statutory
interpretation is still complicated—and fascinating.

Suggested Reading
<< Achtenberg, “With Malice Towards Some.”
<< Elhauge, Statutory Default Rules.
<< Macey and Miller, “The Canons of Statutory

Construction and Judicial Preferences.”
<< Manning, “The Absurdity Doctrine.”
<< Posner, “Statutory Interpretation.”
ÞÞ Can you think of interpretive principles that you apply in
everyday life to give meaning to ambiguous phrases?
ÞÞ Should judges always interpret statutes to avoid absurd

results?

Lecture 8
HOW DO COURTS
REALLY INTERPRET
STATUTES?
S
ometimes, the Supreme Court enforces the plain meaning of a statute’s
text, even though it leads to seemingly unfair results. Other times, the
Supreme Court implements the purpose of the statute notwithstanding
the plain meaning of the text. Still other times, the court does something
in between. This lecture tries to make sense of that by considering these
two questions: Is the enterprise of statutory interpretation hopelessly
incoherent and unpredictable? Is it possible to articulate a theory of statutory
interpretation that explains what courts actually do in cases involving disputes
over the meaning of statutes?
BACKGROUND ON UNITED STATES V. MARSHALL
To frame these questions, it is useful to consider the case of United States v.

Marshall. For background, the federal Controlled Substances Act makes it
a crime knowingly or intentionally to manufacture,
distribute, or dispense certain dangerous drugs,
or to possess one of those drugs with the intent
to distribute it. The statute then specifies
criminal penalties for offenses
involving various drugs. It
establishes minimum sentences
based on the particular drug at
issue and the amount of the
drug at issue.
For example, the statute requires a jail sentence of at least 5 years for someone
convicted of selling between 100 and 999 grams of heroin, and a sentence of at
least 10 years for selling at least 1 kilogram (or 1,000 grams). The penalty for the
sale of between 500 and 4,999 grams of cocaine is at least 5 years in jail; the sale
of 5 kilograms or more draws a punishment of at least 10 years.
These provisions of the statute are known as the mandatory minimums

because they require a person convicted of a particular drug offense to go to
jail for at least some minimum term. As the examples suggest, the mandatory
minimum sentences are a function of two factors: which drug you were caught
with, and how much of the drug you had.
STANLEY MARSHALL
In the case of United States v. Marshall, Stanley Marshall was arrested and
charged with distributing LSD, a hallucinogenic drug. He was convicted for
distributing more than 10 grams of the drug and sentenced to 20 years in
prison. He challenged the length of his sentence.
Here is what the Controlled Substances Act says about crimes involving the
distribution of LSD: a person convicted of selling 1 gram or more of a “mixture
or substance containing a detectable amount” of LSD shall be sentenced to at
least 5 years in prison. A person convicted of selling 10 grams or more shall be
sentenced to at least 10 years. Marshall was convicted of distributing enough
LSD for almost 12,000 doses; by weight, it was over 100 grams. He obviously
was not very happy to be going to prison for so many years, but it would
appear on the surface that the amount that he was selling clearly triggered
that sentence.
MARSHALL’S CHALLENGE
Marshall’s challenge had to do with the way that LSD is distributed,

administered, and ingested. A typical dose of LSD weighs only about 0.05
milligrams. That is almost nothing. LSD is a very potent drug. At that
weight, 20,000 doses of pure LSD together weigh only 1 gram. That raises the
question: How could Marshall’s 12,000 doses have weighed over 100 times
that amount?
Lecture 8–How Do Courts Really Interpret Statutes? 59

The answer is that because a typical dose of LSD is so tiny—almost invisible
to the naked eye—it is usually distributed to consumers on something
else. Pure LSD is typically dissolved in some kind of solvent, like alcohol,
and then sprayed or eye-dropped onto blotter paper, gelatin, or sugar
cubes. A person who takes a dose of LSD takes a small square of the paper
(or the gelatin or the sugar cube) and either swallows it or drops it into
a beverage and drinks it.
The actual LSD in the 12,000 doses that Marshall was convicted of selling
weighed considerably less than 1 gram, not enough to trigger the 5-year
mandatory minimum sentence, let alone the 10-year one. However, he
sold the LSD, as do most dealers, on blotter paper. Although that paper is not
particularly heavy, the 12,000 doses of LSD on paper weighed over 113 grams—
more than 160 times the weight of the pure LSD alone.
Marshall said that he did not sell nearly enough LSD, by weight, to trigger the
mandatory-minimum-sentence provisions of the statute. If this were the right
way to read the statute, then Marshall’s sentence would have been reduced to
a fraction of the one he had received after trial.

MARSHALL’S CASE IN COURT
The legal issue was whether, in calculating the sentence for a convicted
LSD dealer, a court should consider only the weight of the active drug
or instead should include the weight of the carrier medium for the drug,
too. The issue was important and complicated enough that the entire
United States Court of Appeals for the Seventh Circuit decided to review
Marshall’s conviction and sentence, rather than just the ordinary panel of
three judges.
Judge Easterbrook, a former law professor, wrote the opinion for the court,
upholding the defendant’s sentence. He concluded that it’s just “not possible
to construe the words” in the statute “to make the penalty turn on the net
weight of the drug rather than the gross weight of carrier and drug.” He
reasoned that the statute “speaks of a ‘mixture or substance containing
a detectable amount’ of a drug.” He said that the phrase “detectable amount”
is the opposite of “pure.”
After all, if a solution were pure LSD, it would by definition have more than just
a “detectable amount” of the drug. He said that the implication of the phrase
in the statute is that the “mixture” should not “be converted
to an equivalent amount of pure drug.”
Judge Easterbrook conceded that it does not

necessarily follow from this reading that
blotter paper counts as part of a “mixture
or substance” containing LSD within the
meaning of the statute. After all, the phrase
“mixture or substance” cannot possibly
include all carriers.
However, Marshall argued that LSD and blotter

paper are different. He argued that LSD sits on blotter
paper the way that oil floats on water—that is, it doesn’t
actually mix with the paper. As a consequence, he said that the resulting
product cannot count as a mixture under the statute.

Judge Easterbrook responded by saying that that is not exactly what
happens when one puts LSD on blotter paper. LSD is made into a solvent by
putting it in alcohol, and then the solvent is applied to the paper. The fibers
of the paper absorb the solvent. Then, the solvent evaporates, and a tiny amount
of LSD remains. The LSD is in the paper, not on it, constituting a mixture.
Judge Easterbrook made two arguments based on the plain meaning of the
text of the statute. First, he said that the plain meaning of the phrase “mixture
or substance containing a detectable amount” of LSD suggests that something
more than the weight of the pure drug counts. Second, he said that the ordinary
meaning of the term mixture embraces LSD applied to blotter paper.
OTHER DRUGS
Judge Easterbrook also pointed to the mandatory minimum provisions in the

statute for other drugs. For example, there was no dispute that a court would
consider the total weight of cocaine mixed with other material in deciding
how much cocaine a defendant had sold, even though the provision about
cocaine used the same “mixture or substance” language. In fact, the provisions
governing all of the other drugs that triggered mandatory minimum sentences
used the same language—“mixture or substance containing a detectable
amount” of the drug—except one.

The provision for phencyclidine—commonly called PCP—required a sentence
of at least 10 years to any person who distributes “100 grams or more of PCP, or
1 kilogram or more of a mixture or substance containing a detectable amount
of PCP.” For PCP, in other words, Congress distinguished the pure drug from
a mixture or substance containing the drug, requiring a lower weight of the
pure drug to trigger the mandatory minimum sentence.
The defendants in the case were asking the court to read the phrase “mixture
or substance containing a detectable amount” of LSD to mean “pure drug”
for LSD, even though the exact same phrase evidently meant something
different for all of the other drugs on the list. Additionally, they were asking
the court to read the phrase to mean “pure drug” even though, for one of the
other drugs on the list, Congress had actually explicitly distinguished between
mixtures, on the one hand, and the pure drug, on the other. Judge Easterbrook
said that it would stretch the statute’s language past the breaking point to
read it this way.
THE DISSENTS
Judge Cummings dissented. He pointed out that, at the time of the decision,
Congress was in the process of considering a bill that would have amended
the mandatory-minimum-sentence provisions. The amendments would have
made clear that the weight of the carrier does not count for determining the
weight of the mixture or substance.
Judge Easterbrook responded, however, by noting that “subsequent debates are

not a ground for avoiding the import of enactments.” Although the views of
members of a subsequent Congress are “entitled to respect,” ongoing debates
are different from actually enacted statutes; only the latter carry the force of
law. Congress never got around to enacting the amendment.
Judge Posner, another former law professor, wrote a separate dissent. He

began by noting that the sentencing scheme in the statute, which is based on
weight, “works well for drugs,” like cocaine, “that are sold by weight.” For
those drugs, the “dilute form is the product,” and it is natural “to punish its
purveyors according to the weight of the product.”

However, he noted that LSD is usually sold to the consumer “by the dose;
it’s not cut, diluted, or mixed with something else.” Additionally, LSD is
“incredibly light.” Judge Posner also argued that “a quart of orange juice
containing one dose of LSD is not more, in any relevant sense, than a pint
of juice containing the same one dose, and it would be loony to punish the
purveyor of the quart more heavily than the purveyor of the pint.”
Among other arguments, Judge Posner also compared the scheme for LSD,
under the majority’s approach, to the scheme for other drugs under the
statute. Marshall had sold almost 12,000 doses of LSD. That is a lot, but Judge
Posner noted that to get a comparable sentence for selling heroin—a more
dangerous and socially destructive drug—you would have to sell 2 million
doses. The numbers were similar for cocaine.
Additionally, he thought that the statutory scheme as written was irrational.

He thought that the most plausible explanation for the puzzling treatment of
LSD, especially in light of the statute’s treatment of PCP, was that “Congress
simply didn’t realize how LSD is sold.”
TWO VIEWS
Judge Posner wondered what judges could do about this problem. He thought
that the answer “lies in the shadow of a jurisprudential disagreement.” The
disagreement was over the judges’ role in interpreting and applying the law.
One view, which he called the “severely positivistic view,” is that the “content
of law is exhausted in clear, explicit, and definite enactments by legislatures.”
That is the view that Judge Easterbrook implicitly advanced in the case.
On this view, the law is defined by the plain meaning of the text, and usually
nothing more.
The competing view, which Judge Posner called the “legal pragmatist’s view,”
is that the “practice of interpretation and the general terms of the law authorize
judges to enrich positive law with the moral values and practical concerns of
civilized society.” On this view, the judge does more than merely “interpret”
the meaning of the words in the statute; the judge acts as a partner of the
legislature to bring some rationality to the scheme. Judge Posner conceded
that “neither approach is entirely satisfactory.”

The problem with the first view is that when judges treat the plain language as
the law, regardless of consequences, we will inevitably have many cases with
unfair outcomes. The problem with the second view is that if judges can depart
from statutory language to produce fair results, they are really just acting as
legislators—even though people do not get to vote for (or against) them. And
people regulated by the law will never really know what the law requires until
it is too late—until, that is, a judge has applied it.
Judge Posner observed, “our legal system oscillates between the approaches.”
However, he thought that the right approach was clear. He said that “the
positivist view, applied unflinchingly to this case, commands the affirmance of
prison sentences that are exceptionally harsh by the standards of the modern
Western world, dictated by an accidental, unintended scheme of punishment
nevertheless implied by the words of the relevant enactments.”
The pragmatic approach, in contrast, “leads to a freer interpretation, one

influenced by norms of equal treatment.” He thought the best approach was
to interpret the phrase “mixture or substance containing a detectable amount
of [LSD]” to exclude the carrier medium—in this case, the blotter paper, and
in other cases, the orange juice. He thought that the blotter paper and juice
were better viewed as “carriers, like the package in which a kilogram of cocaine
comes wrapped or the bottle in which a fifth of liquor is sold.”
Read this way, the punishment schedule for LSD would make perfectly good
sense and be consistent with the statutory design. Judge Posner reasoned that
Congress was probably just using weight as a proxy for doses of drugs. If so,
then it makes sense to base the punishment for the sale of LSD on the weight
of the pure drug after all.
CONCLUSION
Various touchstones of interpretation were at play in Marshall. Judge

Easterbrook relied on the plain meaning of the text of the statute, using
a textualist approach. Judge Cummings, in contrast, asked how Congress
intended to treat drug crimes involving LSD. He used an intentionalist
approach to the statute.

Both of these approaches are based on the premise that courts are faithful
agents of the legislature—that is, they proceed from the assumption that
a judge’s job is to implement the instructions that the legislature gave,
not to create a new set of instructions. Judge Posner proposed a different
understanding of the judge’s role. He urged the court to engage in what he
has called “pragmatic interpretation” to make sense of the statutory scheme.
Judge Posner said that judges should discard the fiction that they’re seeking
to discern the legislature’s actual purpose. Instead, he said, judges can and
should candidly act like equal participants in the process of creating law.
At least as a formal matter, Judge Posner’s approach is not the majority

approach today. Most theories of statutory interpretation—and most judicial
opinions seeking to interpret statutes—self-consciously adopt the faithful
agent theory.
Suggested Reading
<< Calabresi, A Common Law for the Age of
Statutes.
<< Easterbrook, “Text, History, and Structure in

Statutory Interpretation.”
<< Eskridge, Dynamic Statutory Interpretation.
<< ———, “Public Values in Statutory

Interpretation.”
<< Nelson, “What Is Textualism?”
<< Posner, “Legal Formalism, Legal Realism,

and the Interpretation of Statutes and the
Constitution.”
<< ———, “Statutory Interpretation.”

ÞÞ Does it make more sense to think about statutory
interpretation as a theory about the meaning of language
or instead as a theory about the appropriate judicial role?
ÞÞ Should judges have power to revise statutes to ensure that

they are sensible and workable?

Lecture 9
FEDERAL AGENCIES
AS REGULATORY BODIES
M
uch of this course has been about legislation—that is, how laws are
made and how courts interpret them once they are enacted. However,
this is also a course about regulation. This lecture looks at how the
United States federal government regulates as well as the relationship between
legislation and regulation.
ENFORCING RULES
There are several different mechanisms for enforcing the rules that legislatures
create. First, statutes can be enforced in lawsuits between private parties.
Second, laws can be enforced in criminal prosecutions initiated by the
government against private parties.
This lecture focuses on the third principal means for enforcing the rules that
legislatures announce by statute: agency enforcement. Examples of agencies
include the Equal Employment Opportunity Commission (or EEOC), the
Federal Trade Commission (or FTC), and the Department of the Interior,
to name a few.
Sometimes, agency enforcement of federal statutes is straightforward and

familiar. For example, imagine that one giant company attempts to merge
with another giant company, under circumstances that would give the
resulting company monopoly power. The attorney general might respond by
authorizing Department of Justice lawyers to sue to prevent the merger. When
the department does so, it enforces the federal statute—called the Sherman
Antitrust Act—designed to prohibit the creation of monopolies.
But agencies do much more than that. Some of the things that agencies do
are designed to encourage compliance with laws that Congress has passed,
without actually carrying the force and effect of law. Alternatively, an agency
might gather information or conduct an investigation about some policy issue
and then publicize the results.
Agencies also take actions that have binding legal effect. First, an agency
might act like a court, adjudicating a case involving an alleged violation of
the statute the agency is charged with administering. Agency adjudication
can be more informal, too. For example, a power plant might apply to the
Environmental Protection Agency (EPA) for a permit to use some industrial
process that produces pollution. The EPA will review the evidence submitted
by the plant and then decide whether to grant the permit.
Additionally, agencies often issue rules that carry the force of law. For example,
if the EPA concludes that a particular industrial process always releases an
unacceptable amount of pollution into the air, it can issue a rule that prohibits
the process. When agencies do this, they are acting like a legislature—making
policy and announcing binding rules.
Lecture 9–Federal Agencies as Regulatory Bodies 69

Though there are important constraints on agency authority, federal agencies
often make really important decisions of policy, with implications both for the
parties who are burdened by the policies and the parties who benefit from the
polices. Giving so much authority to federal agencies raises several questions
of constitutional dimension, chief among them: Should we permit Congress
to give agencies in the executive branch the power to decide important
questions of policy?
WHITMAN V. AMERICAN TRUCKING ASSOCIATIONS
In thinking about this question, consider a 2001 Supreme Court decision in

a case called Whitman v. American Trucking Associations. The case involved
a constitutional challenge to a rule that limited the amount of particulate
matter that can be emitted into the air. Particulate matter, sometimes called
soot, is emitted by diesel engines and some industrial processes. To comply
with the rule, diesel engines would have to get cleaner. A trade association of
trucking companies sued because the rule would make it more expensive for
the companies to do business.
The case involved a provision of the Clean Air Act, which Congress first
enacted in the 1960s and then revised in 1970. The section of the law at issue
required “national ambient air quality standards” for air pollutants. The rule
at issue in the case established those standards for particulate matter.
There was little doubt, under the modern understanding of federal power, that
Congress had authority to regulate the amount of pollution that industrial
processes or truck engines are permitted to emit into the air. As a consequence,
the plaintiffs in the case didn’t contend that Congress lacked power to regulate
air pollution.
The rule at issue in the case had been issued by the Environmental Protection
Agency, pursuant to authority that Congress had given to the agency in the
Clean Air Act. The law directed the agency to issue the air quality standards
for pollutants. The agency had promulgated the rule at issue in the case, about
particulate matter, pursuant to that authority.

The trucking companies argued that Congress had impermissibly delegated
the authority to regulate pollution to the EPA. They said, if Congress wants
to make rules to regulate air pollution, then it has to make them itself, rather
than delegating the task to an agency.
In pressing this argument, the trucking companies invoked the nondelegation

doctrine. Under that doctrine, Congress may not delegate the “legislative
power” to someone else. Article I, Section 1 of the Constitution—the
very first provision in the Constitution after the preamble—says that “All
legislative Powers herein granted shall be vested in a Congress of the United
States.” Because those powers are vested in Congress, the theory goes, only
Congress can exercise them. The trucking companies argued that when
Congress gave the EPA power to decide how much air pollution was allowed,
Congress effectively—and impermissibly—gave its legislative power to
the agency.

POWER ISSUES
A claim of impermissible delegation is different from most other types

of separation-of-powers problems. Usually, a separation of powers claim
asserts that one branch has tried to take for itself powers assigned to another
branch. But a claim about delegation is different; it asserts that Congress has
impermissibly tried to give away its power to someone else.
However, it is not obvious that we should read the language of Article I to mean
that Congress cannot delegate the legislative power. After all, the Constitution
says something about the president and executive power that is very similar
to what it says about legislative power.
Article II of the Constitution begins by saying that the “Executive Power shall
be vested in a President of the United States.” Still, it is widely accepted that
the president can delegate authority to other officials, including members of
the cabinet.
This begs the question: If power is understood to be delegable, then why

isn’t legislative power delegable, too? There are two main reasons why the
Supreme Court has said that Congress cannot delegate the legislative power.
First, the nondelegation doctrine is grounded in political theory. The elected
legislature is politically accountable to the voters. As a result, it should make
the important policy decisions that affect voters’ lives.
Second, differences between executive and legislative power might justify

different rules about whether the power can be delegated. The essence of
legislative power is the authority to make policy. However, essence of executive
power is the authority to enforce the law, not to make the law. In other words,
there is a qualitative difference between legislative power and executive power.
Legislative power is the power to make policy; executive power is just the
power to enforce those policy choices.
The basic intuition of the nondelegation doctrine is this: Congress has the
exclusive power to make law, which means the power to make important
policy decisions. As a consequence, Congress cannot just give the power to
make policy to someone else.

WHITMAN V. AMERICAN TRUCKING ASSOCIATIONS,
CONTINUED
Although as a formal, theoretical matter the Supreme Court recognizes the

nondelegation doctrine, in practice the court doesn’t aggressively apply it. In
practice, Congress can delegate substantial policy-making authority to agencies.
To see why, this lecture returns to Whitman v. American Trucking Associations.
In that case, the court explained that Congress can delegate some authority to
agencies as long as it doesn’t actually delegate the legislative power—that is, the
power that the Constitution gives exclusively to Congress. If Congress makes
the fundamental policy choices and then gives an agency some discretion
about how to enforce those choices, Congress hasn’t impermissibly delegated
the legislative power; instead, it has exercised it, leaving only executive power
for the agency.
The challenge is how to tell the difference between a delegation of the legislative
power on the one hand and a delegation of the mere discretion to enforce
the law on the other. Sometimes we have an intuition about the difference.
For example, when a prosecutor declines to prosecute someone who merely
possessed a small quantity of drugs, she is exercising discretion about how to
enforce the law, not making the law.
In Whitman, the court put it another way. The court said that when Congress
gives authority to an agency, it must announce an “intelligible principle” with
which the agency must conform. If Congress does so, the court will hold that
the agency’s exercise of the delegated power doesn’t constitute an exercise of
the legislative power.
In this view, Congress had made the difficult policy choice between public
health and cost to industry. The agency had to fill in the details, of course, but
no one seriously argued that Congress had to determine the exact permissible
exposure level for every conceivable pollutant. Congress made the fundamental
policy choice, and all it left for the agency was the implementation of the law.

Even though it received a lot of attention, Whitman v. American Trucking
Associations turns out to have been an easy case. The court resolved it
unanimously.
THE NONDELEGATION DOCTRINE
Some questions remain: Why has the Court been unwilling to police
congressional delegations of policy-making power? Why doesn’t the
nondelegation doctrine have any teeth?
In thinking about those questions, consider a different example involving

public health. Congress has charged the Occupational Safety and Health
Administration (or OSHA) with the responsibility to issue standards for the
permissible exposure of workers to toxins in the workplace. Under the statute,
for each toxin, the agency must choose the standard that “most adequately
assures, to the extent feasible, on the basis of the best available evidence, that
no employee will suffer material impairment of health,” even after consistent
exposure to the toxin.
In the 1970s, a trade group for the petroleum industry challenged the agency’s
attempt to set exposure limits for benzene, which causes cancer at sufficiently
high doses. Was the standard in this statute a broader delegation of authority
than the one at issue in Whitman v. American Trucking Associations?

The main difference was that the workplace statute required the agency to
choose the level that most adequately assures health “to the extent feasible.” The
meaning of this phrase wasn’t entirely clear. The petroleum industry argued
that it required the agency to balance any gains in worker health against the
costs to industry of complying with the rule. Others argued that the phrase
merely referred to the agency’s obligation to make determinations in light of
imperfect evidence.
COSTS AND BENEFITS
Assume that the statute, properly interpreted, meant that the agency had
discretion to decide whether to consider costs on employers in setting the
correct standards. In other words, imagine that Congress had given the
agency the power to decide whether to do cost-benefit analysis before setting
exposure limits. If that were the case, should a court strike the statute down
as an impermissible delegation of authority? Would that statute contain an
intelligible principle?
Whether an agency should consider the costs of a proposed standard, or instead

should simply prioritize health over all other concerns, is a very important
question of policy. The decision whether to consider costs on employers—costs
that they will pass on to consumers—in seeking to protect employee health
has enormous consequences, both for the economic viability of thousands of
employers and for the health of millions of employees. If the purpose of the
nondelegation doctrine is to ensure that Congress makes the difficult policy
determinations rather than leaving them to agencies to resolve, then there
is a strong case that this hypothetical statute would be an impermissible
delegation of legislative power.
In this example, the decision to regulate a particular toxin—and the permissible

level of exposure that the agency selects—can have profound effects both on
worker health and the financial well being of huge segments of the economy.
Such decisions are, in practice, incredibly important. This insight helps to
illustrate why the court hasn’t applied a robust nondelegation doctrine, and
instead has permitted very broad delegations of policy-making authority to
administrative agencies.

First, it doesn’t make sense to require Congress to make every conceivable
regulatory decision because Congress cannot, in enacting a statute, anticipate
every question that is likely to arise or every change of circumstance that might
occur in the future. Second, there is some virtue in giving decision-making
authority to agencies that have developed expertise in a particular area of
regulation. For instance, OSHA and the EPA have hundreds of scientists and
other policy experts on staff, and it is their job to consider the best available
evidence about health and related concerns.
Third, and perhaps most important, the distinction between legislative and
executive power that serves as the basis for the nondelegation doctrine—that is,
the distinction between making the law and enforcing the law—in practice is
quite elusive. Faced with these sorts of questions, judges have recognized that
it is often not possible to differentiate in any principled way between making
the law and enforcing the law. As a result, the court won’t strike down a statute
under the nondelegation doctrine unless there is absolutely no intelligible
principle to guide the agency’s exercise of discretion.
Suggested Reading
<< Gailmard and Patty, Learning While
Governing.
<< Hamburger, Is Administrative Law Unlawful?
ÞÞ Should we insist that Congress make all important policy
choices? Alternatively, should Congress be permitted to
delegate that power to executive agencies?
ÞÞ How do we tell the difference between the power to make

important policy choices and the mere discretion to fill in
the gaps in a statutory scheme?

Lecture 10
POLITICAL CONTROL
OF AGENCY DECISION
MAKING
T
he United States Congress often gives agencies very broad power to make
policy. However, members of Congress are still interested in what the
agencies do with that power. This lecture considers the forms of control
that Congress exercises over agency decision making. Other topics include the
president’s power over those agencies and how regulation by federal agencies
raises questions of constitutional importance.
INS V. CHADHA
An interesting example is a Supreme Court case from 1983 called INS v.

Chadha. The dispute arose when the Immigration and Naturalization Service
(INS) decided to deport a man named Jagdish Chadha, who had overstayed
his student visa.
In the Immigration and Nationality Act, Congress had provided a safe harbor
for aliens facing deportation. If the person could show that he had resided in
the country for at least seven years, was of good moral character, and would
face extreme hardship if deported, the government could decide to suspend
his deportation.
At the time, the INS was a division of the Department of the Justice,
so the attorney general had the final say over whether to suspend an
alien’s deportation. The attorney general decided that Chadha had met the
requirements under the statute and suspended Chadha’s deportation.
However, the House of Representatives passed a resolution disapproving the
suspension. Chadha challenged this decision. The Supreme Court concluded
that this so-called legislative veto was unconstitutional. As a consequence, the
House’s decision was invalid, and Chadha got to remain in the United States.
Congress had included the legislative veto provision in the statute—as it had
in hundreds of others—because it wanted to have some control over the way
that an agency exercised the power that Congress had given it. However, the
Constitution says that before a bill, resolution, order, or vote can become law,
it must be passed by both houses of Congress and then signed by the president.
Alternatively, if the president vetoes it, is must return to the Congress and
passed by two-thirds of each house.
The requirement that both houses of Congress approve a measure before it

becomes law is known as the bicameralism requirement. The requirement
that all such measures be presented to the president for a signature or veto is
known as the presentment requirement.
The court reasoned that the legislative veto violated these requirements.
The veto had the effect of law because it changed Chadha’s legal rights. But
the Senate had not voted to disapprove the Attorney General’s decision, and the
House’s veto was not presented to the president. Accordingly, the court said,
it was unconstitutional.
Justice White dissented in the case. He thought it was strange that agencies
can make rules with the force of law while Congress, which explicitly enjoys
the legislative power, cannot exercise a veto to prevent those rules from
taking effect.

OTHER MEASURES
Though Congress cannot use the legislative veto to control an agency’s

decisions, it does have other measures it can use when it does not like an
agency’s actions. First, the most straightforward form of control is for Congress
to pass a new law that reverses the decision or makes clear that the agency’s
view is no longer permissible. This approach is difficult, however, because of
the challenges present in passing new laws.
The second measure is that when Congress creates a new program, it can make
the law more specific, delegating less power to the agency to make important
decisions. Third, Congress can threaten to cut off funding from agencies that
make decisions Congress doesn’t like. Fourth, Congress can use the power of
oversight; it can summon the head of the offending agency before one of its
committees and rake that official over the coals.
Fifth, Congress can use its power over the appointment of agency officials.
Presidential nominations of important officers are subject to the advice and
consent of the Senate; the Senate can refuse to confirm officials who are likely
to interpret statutes contrary to congressional intent.
AGENCY INDEPENDENCE
Sometimes, a regulatory agency enjoys some degree of independence

from political control. In fact, in one sense, every agency is independent
from political control. For example, if the administrator of the EPA issues
a rule we don’t like, we can’t vote her out of office.
That does not mean that those agencies are completely unaccountable to the
voters. The president plays an important role both in deciding who will serve
at the agencies and in dictating the policy agenda that those agencies will
pursue. If we disagree with something that an agency does, we can vote for
someone else for president. Of course, few voters decide which candidate to
vote for based on a rule the EPA issued. But plenty of people base their votes
on the candidates’ more general approach to regulation.
Lecture 10–Political Control of Agency Decision Making 79

This theory of accountability depends at least in part on the assumption that
the president has complete discretion to decide who will serve in important
positions and to dictate the agenda that they will pursue. This assumption
isn’t entirely correct. The president’s power to control agency decision-making,
like Congress’s, is incomplete.
Some of the reasons are practical rather than legal. The federal government
is huge, and it is just not feasible for the president to exercise total, day-to-day
control over its activities. There are also some norms that are deeply entrenched
even if not constitutionally required, such as the norm that the president
should not interfere in decisions about whom to prosecute.
There are also legal constraints on the president’s power to control agency
decision-making. For example, even if Congress cannot exercise total control
over the agencies, it can still try to limit the president’s control. For instance, if
Congress wants to constrain the president’s ability to control policy-making,
it can limit the president’s power to fire the officials in charge of the
decision-making.

THE PRESIDENT’S ABILITY TO FIRE OFFICIALS
There is a specific clause in the Constitution to govern the process for

appointing government officials, but other than giving Congress the power
of impeachment, the document is silent on the power of removal.
The Supreme Court didn’t confront the removal power until the 20th century.
In 1917, President Wilson appointed Frank Myers to serve as the postmaster
of Portland, Oregon. Less than three years later, Wilson decided to fire
Myers. But a federal statute enacted in 1876 provided that postmasters would
serve for four-year terms. It also said that postmasters “shall be appointed
and may be removed by the President with the advice and consent of the
Senate.” The Senate did not take any action. Myers grudgingly gave up his
office, but he sued to recover his salary for the portion of his four-year term
that he didn’t get to serve.
The Supreme Court concluded that Myers wasn’t entitled to his salary because
the law impermissibly interfered with the president’s power to remove him.
The court began by noting that the Constitution vests the executive power—
and thus the power to execute the laws—in the president.
The court reasoned that the president can’t meaningfully ensure that the law
is faithfully executed if he can’t control the officials who actually exercise
executive power. The court concluded that the president must have unfettered
authority to fire executive officials.
The case was a bit more complicated, however, because Myers was an “inferior
officer.” The Appointments Clause of the Constitution distinguishes between
officers and inferior officers. It says that Congress can let someone other than
the president appoint inferior officers, which means that they can be subject
to the direct control of someone other than the president. In another case, the
court had held that Congress can limit the president’s power to fire inferior
officers as a way to prevent undue political patronage in the civil service.

Notwithstanding its broad language about the president’s power, the court
in Myers v. United States conceded that Congress can sometimes limit the
president’s power to fire an inferior officer by vesting the power to remove
in a principal officer, rather than directly in the president. However, the
court said that the statute governing the removal of the postmaster was still
unconstitutional, because Congress cannot reserve for itself the power to
remove an executive official.
HUMPHREY’S EXECUTOR V. UNITED STATES
Only nine years later, the Supreme Court advanced a very different vision of
the president’s power in a case called Humphrey’s Executor v. United States.
President Hoover had appointed William Humphrey to serve a seven-year
term on the Federal Trade Commission, an agency charged with enforcing
the laws prohibiting unfair trade practices.
Two years later, President Roosevelt fired Humphrey.

Humphrey sued for his salary, arguing that the
president lacked power to fire him because the
statute creating the commission said that
the president could fire a commissioner
only for “inefficiency, neglect of duty, or
malfeasance in office.” President Roosevelt
didn’t claim that Humphrey’s work satisfied
that standard; instead, he fired him because,
as he said at the time, they didn’t see eye to
eye on policy.
Unlike the postmaster in Myers v. United States,

who was an inferior officer, commissioners of the FTC are
principal officers. The reasoning in Myers seemed to require the conclusion
that the president has unfettered authority to remove any principal officer
charged with executing the law. However, the court in the Humphrey’s
Executor v. United States upheld Congress’s restriction on the president’s
removal power.

AN ONGOING DEBATE
The debate over whether Congress can limit the president’s control over
officials charged with executing the law has been going on since the founding of
the country. For example, there are clauses in the Constitution that contemplate
officials exercising authority beyond the absolute control of the president.
For instance, the Opinion Clause in Article II says that “The President … may
require the Opinion, in writing, of the principal Officer in each of the executive
Departments, upon any Subject relating to the duties of their respective
Offices.” This clause would be redundant if all executive officers were subject
to the president’s unfettered control.
If the president can fire executive officials for any reason at all, let alone for
having a policy agenda that conflicts with the president’s, then of course
the president can require them to give him their opinions on issues relating
to their duties. This clause implies not only that Congress has authority to
determine the duties of executive officers, but also to vest them with power
exercised independent of presidential control.
INDEPENDENT PROSECUTORS
Do restrictions on the president’s power to fire officers promote a functional

system of government where crass political considerations do not taint the
work of governing? Or do they create a dangerous “fourth
branch” of government, not directly accountable
to anyone?
In thinking about this question, consider

the problem of independent prosecutors.
While he was still president, Richard
Nixon had permitted his attorney general
to appoint a special prosecutor to investigate
wrongdoing by the president and close
campaign associates in connection with the
Watergate scandal.

When it appeared that the prosecutor might uncover serious crimes, Nixon
ordered the attorney general to fire him. In the infamous Saturday Night
Massacre, the top two officials at the Department of Justice resigned rather
than fire the special prosecutor. Nixon prevailed upon Robert Bork, then the
solicitor general, to fire him.
Nixon eventually relented under extreme political pressure and allowed the
appointment of a new special prosecutor. The new prosecutor pursued the
investigation, and Nixon ultimately resigned.
However, the episode revealed the conflict of interest inherent in our system.
The president controls the Department of Justice through his power to fire the
attorney general. If there is credible evidence to suggest that the president or
a high-ranking executive official has committed a crime, how do we ensure
that it’s investigated properly?
After Nixon’s resignation, Congress enacted the Ethics in Government Act,

which created a process for the appointment of independent counsels to
investigate executive wrongdoing. The statute required the attorney general
to ask a panel of judges to appoint an independent counsel if there were
reasonable grounds to believe that a high-ranking official violated the law.
Once appointed, the attorney general could fire the independent counsel
only for “good cause,” which limited the president’s power to interfere in the
investigation.
In other words, independence from presidential control was Congress’s

way of responding to the problem of conflict of interest. But in a 1988 case
called Morrison v. Olson, a government official being investigated by an
independent counsel for lying to Congress challenged the constitutionality
of the independent counsel law, arguing that independence was the fatal flaw
in the scheme.
The court upheld the independent counsel provisions of the law. It read
the Myers and Humphrey cases together to mean that congress can limit
the president’s power to fire an officer as long as the limitation doesn’t
impermissibly interfere with the president’s ability to perform his
constitutional duties.

Justice Scalia was alone in dissent. He advanced the unitary executive theory,
arguing that Congress cannot impose any limits on the president’s power to
fire an officer charged with executing the law.
Suggested Reading
<< Aberbach, Keeping a Watchful Eye.
<< Calabresi and Rhodes, “The Structural

Constitution.”
<< Kagan, “Presidential Administration.”
<< Lessig and Sunstein, “The President and the

Administration.”
ÞÞ Should Congress be permitted to veto executive action
that it finds misguided?
ÞÞ Should Congress be permitted to limit the president’s

power to fire an official in the executive branch?

Lecture 11
JUDICIAL REVIEW
OF AGENCY RULINGS
F
ederal agencies have substantial power to make policy in the US political
system. Congress has given them authority to make rules about all sorts
of important matters—like environmental protection, traffic safety,
health care, and more. Yet agencies and the officials who run them are not
directly accountable to the voters. Instead, the principal mechanism in the
US system for keeping agencies in check is judicial control. Courts routinely
review the decisions of agencies to ensure that they are not exceeding the
authority that Congress has given them.
This lecture considers two basic forms of control that courts exercise over
agencies. First, courts review agency decisions to ensure that they are
procedurally sound. Second, courts review agency decisions to ensure that they
are well reasoned and based on relevant and appropriate considerations.
JUDICIAL REVIEW OF AGENCY PROCEDURES
In 1972, Congress created the Special Supplemental Food Program for

Women, Infants, and Children (known as the WIC program). The program
was designed to ensure that pregnant women, infants, and young children in
families with inadequate income get adequate nutrition. Under the program,
the Department of Agriculture gives money to state and local agencies to fund
the purchase of nutritious food.
Congress directed the agency to fund

“those foods containing nutrients known
to be lacking in the diets of populations at
nutritional risk and, in particular, those
foods containing high-quality protein, iron,
calcium, vitamin A, and vitamin C.” Notice,
then, that Congress gave the agency guidance
about what types of food to provide, but
delegated to the agency the power ultimately
to decide which specific foods were appropriate.
Under this standard, the agency authorized funding for milk. In the mid-1970s,
the agency also started to allow the grants to be used to purchase flavored
milk (such as chocolate milk) as a substitute for regular, unflavored milk.
In 1978, Congress reauthorized funding for the program and modified the
agency’s responsibility under the statute. Congress said that the agency should
issue regulations specifying the foods that are available under the program,
and that in doing so, the agency should “assure that the fat, sugar, and salt
content of the prescribed foods is appropriate.”
The agency then gave notice that it was planning to issue a new rule about what
foods would be available under the program. The Administrative Procedure
Act, the statute that provides general rules of procedure for federal agencies to
follow whenever they act, requires an agency that wants to issue a rule to do
three principal things. First, the agency has to publish notice of the proposed
rulemaking. The notice has to provide either the terms of the proposed rule
or a description of the subjects and issues involved.
Lecture 11–Judicial Review of Agency Rulings 87

Second, the agency has to give interested persons “an opportunity to participate
in the rule-making through submission of written data, views, or arguments.”
Third, after considering those comments, the agency has to accompany the
final rule with a “concise general statement” of the rule’s basis and purpose.
COMPETING INTERESTS
The agency rule-making process is designed to ensure that parties who

will be burdened by a regulation and parties who will benefit from it can
help to shape the content of the agency’s rules. Because crafting regulation
requires the balancing of competing interests, we want policy makers to
consider those competing interests.
SPECIALTY FOODS
In the case involving the WIC program, the agency published notice of
a proposed rule to specify foods that are eligible for funding. The proposal
listed milk as one of those foods. The proposal also specifically preserved the
agency’s prior approach, which was to permit the substitution of flavored milk
for regular milk. The notice also explained that the agency was concerned
about high-sugar diets, and it specifically proposed a limit on sugar content
in eligible breakfast cereals.
Many of the comments that the agency received about the proposal
recommended that the agency remove flavored milk from the list of approved
foods because of concern over its high sugar content. The agency listened to
those suggestions, and the final rule excluded flavored milk from the list of
approved foods.
In issuing the rule, the agency acknowledged that it had previously funded
flavored milk, and that the notice hadn’t proposed any change to that approach.
The agency explained that “the comments and the Department’s policy on
sugar validate” changing the final rule to delete flavored milk from approved
food packages.

However, a trade association of companies that make and sell chocolate then
filed suit to challenge the rule. The association argued that it did not receive
adequate notice that the agency was considering removing flavored milk from
the approved list of foods.
The agency responded to the lawsuit by arguing that the notice of the
rulemaking advised the public of its general concern about foods with high
sugar content, which should have given interested parties notice that it would
consider eliminating any food with such content. It also argued that the specific
inclusion of flavored milk in the proposed rule carried the implication that
flavored milk was a subject for consideration and debate.
However, the court concluded that the agency had not given sufficient notice
that it might delete flavored milk from the approved list. It invalidated the
rule, effectively instructing the agency to start the process from scratch.
The court in the chocolate milk case acknowledged that a final rule could
depart from a proposal to reflect comments received. But the court said that
the final rule cannot depart so much from the proposal that interested parties
are denied fair notice that their interests are implicated.
The court said that the final rule must be a “logical outgrowth” of the rule
proposed in the agency’s notice. The court concluded that a rule prohibiting
flavored milk is not a logical outgrowth of a proposal to permit the inclusion
of flavored milk, because the final rule was the exact opposite of the proposed
rule, at least with respect to milk.
The takeaway point is that there are meaningful procedural limits on an

agency’s ability to use the power that Congress delegates to it. The specific
requirements in the Administrative Procedure Act don’t sound all that
demanding; the law requires only that the agency give notice of the proposed
rule, an opportunity for interested parties to comment, and a “concise general
statement” of a rule’s basis and purpose.
The agency did all of those things in the chocolate milk case. Still, as that case
suggests, courts have interpreted these requirements to impose more onerous
obligations on agencies that want to issue rules.

JUDICIAL REVIEW OF AGENCY REASONING
The chocolate-milk case is an example of courts policing agency decisions

for compliance with procedural requirements. That is not the only way that
courts ensure that agency decisions are defensible.
Consider a famous case about automobile safety. In 1966, alarmed by the

number of deaths and injuries caused by car accidents, Congress enacted
the National Traffic and Motor Vehicle Safety Act. The law gave the secretary
of transportation authority to issue safety standards that, in light of available
safety data, would be “reasonable, practicable, and appropriate” and would
“contribute to carrying out the purposes” of the act.
In 1967, the Department of Transportation issued a rule that required the

installation of seatbelts in all cars. The problem was that, at least at the time,
most people did not bother to wear their seatbelts. As a result, the number of
traffic injuries did not go down as much as the agency had hoped.
The agency began consideration of passive occupant restraint systems—that

is, devices that do not require any action taken by the occupant of the car.
At the time, two types of automatic crash protection were feasible: airbags
and automatic seatbelts, which deploy as soon as a person sits in and shuts
the door.
Between 1969 and 1972, the agency issued a series of rules to require at least one
of these passive restraints on all cars. The requirement was to take effect with
all cars manufactured after August 1975. Cars manufactured in the interim
were required to feature either passive restraints or an ignition interlock, which
would prevent the driver from starting the car or sound a continuous buzzer
if the seatbelts were not fastened. Although the car manufacturers challenged
these rules in court, they lost, and the rules took effect.
Because it was cheaper, most car manufacturers chose to install the ignition
interlock system. This feature was incredibly unpopular, and Congress
responded by passing a law prohibiting the agency from requiring or even
permitting car manufacturers to use ignition interlocks or buzzers to indicate
that seatbelts were not in use.

Under pressure from the carmakers, the agency then extended the deadline
to install passive restraints. Then, in 1976, the secretary of transportation
suspended the passive restraint requirement, based on the expectation that
there’d be public resistance to the new systems.
The next year, the secretary restarted the process, issuing a new passive-
restraint rule to require the gradual phasing in of a requirement that all cars use
either airbags or automatic seatbelts. The evidence before the agency suggested
that the requirement would save around 9,000 lives per year. Although the car
manufacturers challenged this rule, too, the courts upheld it.
AIRBAGS AT ISSUE
In 1981, yet another new secretary of transportation rescinded the passive-

restraint rule after giving an opportunity for notice and comment. The agency
did not suddenly doubt the effectiveness of passive restraints, but the safety
data that the agency had relied on in issuing the passive-restraint rule had
assumed that car manufacturers would install airbags in 60 percent of new
cars, with automatic seatbelts in the remaining 40 percent.

By 1981, though, it had become apparent
that carmakers planned to install the
cheaper automatic seatbelts in 99 percent
of cars. This meant that the lifesaving
potential of airbags would not be
realized. In addition, because consumers
strongly disliked automatic seatbelts,
carmakers planned to install belts that
could be easily detached; that made
them similar in practice to traditional,
manual belts, which many drivers had
long ignored.
As a consequence, the agency reasoned

that the safety benefits of the new rule
would end up being quite minimal, but
would cost something like $1 billion for
carmakers to comply with the rule. The agency reasoned that the small safety
benefits did not justify the significant costs. The agency also expressed concern
about the possibility of public resistance to further requirements.
State Farm Insurance Company and others sued, challenging the agency’s
rescission of the rule. They argued that the repeal was “arbitrary and
capricious,” and in violation of the Administrative Procedure Act. The
Supreme Court agreed with the challengers and invalidated the agency’s
decision to rescind the rule.
THE COURT’S REASONING
The Court began by saying that a reviewing court is not supposed to “substitute
its judgment for that of the agency.” However, it then said that when an agency
issues a rule, it has to “examine the relevant data” and provide a “satisfactory
explanation for its action,” including a “rational connection between the facts
found and the choice made.” In the court’s view, the agency’s repeal of the
rule failed to satisfy this standard.

The main problem was that the agency could have responded to the carmakers’
decision to use detachable automatic seatbelts by changing the rule to require
airbags in all cars. The agency did not doubt that airbags were effective devices
for saving lives; it simply reasoned that, because carmakers decided instead
to use a cheaper technology, the hoped-for safety benefits of airbags would
not materialize.
The court said that the “logical response to the faults of detachable seatbelts
would be to require the installation of airbags,” rather than to repeal the rule
entirely. The court also concluded that the rescission of the rule was invalid
because the agency was “too quick to dismiss the safety benefits of automatic
seatbelts.” If nothing else, the court noted that a detachable belt requires some
affirmative action by the occupant of the car to detach it, whereas it takes no
action to ignore a traditional manual seatbelt.
The court’s approach in this case is known as hard-look review because

the court took a hard look at the agency’s reasoning. Unlike the approach
considered in the chocolate milk case, which focused on whether the agency
had followed the right procedures in creating a rule, hard-look review focuses
on the substance of agency decision making. The issue was whether the
agency’s decision was reasonable and made sense in light of the evidence
before the agency.
AFTER THE CASE
After the automobile case, President Ronald Reagan’s Department of

Transportation responded by issuing a new rule. The rule did not require
airbags in all cars, but it did require the phasing in of a passive-restraint
requirement for all cars, which carmakers could satisfy with airbags, enhanced
padding in the car’s interior, or automatic seatbelts.
The rule also provided that the passive-restraint requirement would be

rescinded if, within five years, two-thirds of the population of the United
States was covered by state compulsory-seatbelt-usage laws that met certain
specified conditions, such as a mandatory minimum fine for car occupants
who failed to comply.

The Reagan administration’s hope was that most states would adopt
mandatory-seatbelt-use laws, which would eliminate the requirement that
carmakers install airbags or other expensive equipment in cars. Most states
did so, but after a strategic lobbying effort by groups such as Mothers Against
Drunk Driving, most states carefully crafted laws that would not trigger the
conditions for rescission of the federal rule.
As a consequence, seatbelt usage increased dramatically and carmakers

remained under the obligation to install passive restraints in all cars. A few
years later, in response to public pressure, Congress enacted a law that required
all new cars to have airbags.
Suggested Reading
<< Mashaw and Harfst, The Struggle for Auto
Safety.
<< McGarity, “Some Thoughts on ‘Deossifying’

the Rulemaking Process.”
<< Schuck, Foundations of Administrative Law.
<< Sharkey, “State Farm ‘With Teeth.’”
ÞÞ Should we make it easy for agencies to make policy in

order to address pressing public problems, or should we
find ways to slow down agency policy making?
ÞÞ Do judges have the expertise to review agency rules about

complex and technical matters?

Lecture 12
WEIGHING AGENCY
INTERPRETATIONS
OF STATUTES
T
here are many federal agencies with significant authority to implement
complicated statutory regimes. In implementing and enforcing those
statutes, agencies inevitably make judgments about what those statutes
mean. As a consequence, when a dispute about the meaning of a statute arrives
at court, the judges often are not writing on a clean slate. This lecture looks at
what courts do with an agency’s interpretation.
BACKGROUND ON CHEVRON V. NRDC
The seminal case on

statutory interpretation
is Chevron U.S.A., Inc., v.
National Resources Defense
Council, Inc. Known as
Chevron v. NRDC for short,
the case involved a rather
complicated regulatory
scheme for reducing air
pollution. The Clean Air
Act amendments of 1977
required states that had not
achieved certain air-quality
standards to establish
permit programs to regulate
sources of pollution.
The law said that states could not grant permits for “new or modified major
stationary sources” of air pollution unless several stringent conditions were
met. Essentially, if a factory wanted to install a new piece of machinery, it
had to install a state-of-the-art one, not one that belched out pollution.
The dispute was over what the phrase “stationary source” meant. If the
phrase referred to each device or machine that emitted pollution, then
a company could not get a permit to install or modify any machine in its
factory unless it could show that the machine was the cleanest type available.
But if the “stationary source” was the factory itself, then a company could
replace a machine with another one that did not comply with the newest air-
quality rules as long as the total pollution emitted by the factory remained
constant.
In 1980, President Carter’s Environmental

Protection Agency adopted an environmentalist
view on the topic. It issued a rule that required
a permit whenever a company wanted to
install or modify a piece of equipment
in its factory. This would require factories to
use state-of-the-art equipment every time that
they installed or replaced a piece of machinery
that emitted pollution.
After President Reagan’s election, the EPA issued a new

rule that adopted the industry-friendly view of the statute. The new rule
defined “stationary source” as “all of the pollution-emitting activities that
belong to the same industrial grouping.” In other words, under the rule, the
stationary source was the entire factory. If there were several machines in one
plant that emitted pollutants, the plant could replace one of them with a new
piece that did not comply with the newest air-quality standards as long as the
overall pollution emitted by the plant remained constant.

THE NRDC CHALLENGE
An environmental group called the Natural Resources Defense Council

(NRDC) challenged the rule. The NRDC claimed that the agency’s rule was
inconsistent with the statute. The NRDC said that the phrase “stationary
source” meant a piece of equipment, not an entire factory. In other words, the
NRDC argued that the agency had misinterpreted the statute, and it asked
the court to invalidate the rule as inconsistent with the statute.
However, the Supreme Court said that when a court reviews an agency’s
interpretation of the statute that it administers, the court’s inquiry is different
from the typical case of statutory interpretation. If the court, applying
traditional tools of statutory interpretation, concludes that the statute resolves
the question at issue, then the court should enforce the statute in light of that
meaning. Neither the court nor the agency has authority to ignore Congress’s
clear directions.
The court then said that if Congress has not directly addressed the
precise question at issue, then the court shouldn’t “simply impose its own
construction on the statute, as would be necessary in the absence of an
administrative interpretation.” Instead, the court said, “if the statute is silent
or ambiguous with respect to the specific issue, the question for the court is
whether the agency’s answer” is based on a reasonable interpretation of the
statute.
The court asserted that the court has to uphold the agency’s interpretation
even if it was not the one “the court would have reached if the question initially
had arisen in a judicial proceeding.” Under this approach, the court upheld
the EPA’s rule about stationary sources.
This means that if a statute does not resolve the specific question at
issue in the case, then the court reviewing the agency’s interpretation does
not try to figure out the best, most faithful interpretation of the statute.
Instead, it defers to the agency’s view and upholds that view as long as it is
reasonable.
Lecture 12–Weighing Agency Interpretations of Statutes 97

THE CASE’S IMPACT
Chevron v. NRDC case raised an institutional question: Who should get to

decide what a regulatory statute means? The decision in the Chevron case
had the potential to revolutionize both the law of the regulatory state and the
courts’ role in statutory interpretation. Before Chevron, the courts tended
to consider agency interpretations of ambiguous statutes solely for their
“power to persuade.” If the agency’s view made lots of sense as a matter of
statutory interpretation, then a court might accept it—just as it might accept
a convincing argument about statutory meaning advanced by one of the
lawyers in the case.
But in Chevron, the court said that judges are obligated to accept reasonable
agency interpretations of ambiguous statutes, even if there are compelling
alternative interpretations of the statute.
However, statutory interpretation by judges is alive and well, even after

Chevron, for at least two reasons. First, many cases of statutory interpretation
do not involve agency administration or interpretation. For example,
when Congress passes a law that makes some act a crime, there is no agency
that’s given power to make policy or to implement the statute by regulation.
Courts still decide the meaning of such statutes without having to defer to
any agency.
Second, the Chevron test is a two-step test. Courts are obligated to defer to
agency interpretations only if the statute is ambiguous with respect to the
question presented. When the statute resolves the question at issue, the court
does not defer.
MCI TELECOMMUNICATIONS CORP. V. AT&T CO.
The importance of Chevron turns to a large degree on how willing judges are
to find statutory ambiguity in the first place. Two other cases give a sense of
how courts approach statutory ambiguity in the wake of Chevron.

First, consider a case from 1994 called MCI Telecommunications Corp. v.
AT&T Co. The case involved a challenge to a rule issued by the Federal
Communications Commission. The Communications Act of 1934 authorized
the FCC to regulate the rates charged for communications services to ensure
that they are reasonable. The law required telephone service providers to file
their rates with the agency and to charge only the filed rates to their customers.
The law also said that the FCC had discretion, if it could show good cause, to
“modify any requirement made under the authority” of the statute.
When Congress passed the Communications Act, AT&T had a monopoly on

long-distance service. In the 1970s, technological advances made it possible
for other companies to compete. To help those other companies to create
a competitive market, the FCC issued a rule declaring that the filing of rates
was optional for all “non-dominant carriers,” which essentially meant everyone
except AT&T. Not surprisingly, AT&T challenged the rule, arguing that the
agency did not have power to eliminate the rate-filing requirement.

The FCC defended the rule by arguing that it had simply used the explicit
power in the statute to “modify” the requirement that companies file their
rates with the agency. At a minimum, the agency said, its interpretation of
the statute, which gave it discretion to modify requirements imposed under
the law, was reasonable.
The court disagreed and struck down the rule in a 5-3 decision. The court’s
reasoning was that the word modify “has a connotation of increment or
limitation.” The court then cited several dictionaries that define modify as
“change moderately or in minor fashion.” Accordingly, the court said that
the statute did not give the agency the power to make fundamental changes
to the statutory scheme.
Under Chevron, a court has to defer to a reasonable agency interpretation of

a statute only if the statute does not resolve the question at issue. In the MCI
case, the court concluded that the plain meaning of the word modify prohibited
the agency from waiving the core requirements of the statute. There was no
need to defer to the agency’s interpretation because the statute itself resolved
the question at issue.
FDA V. BROWN & WILLIAMSON TOBACCO CORP.
Another case relevant to the Chevron doctrine is FDA v. Brown & Williamson
Tobacco Corp., which involved the US Food and Drug Administration’s attempt
to regulate tobacco products. The Food, Drug, and Cosmetic Act gives the FDA
authority to regulate “drugs” and “devices.” The Act defines drug to include
“articles intended to affect the structure or any function of the body.” It defines
device as “an instrument, apparatus, implement, machine, or contrivance that’s
intended to affect the structure or any function of the body.”
The act also gives the FDA authority to regulate combination products, which
are a combination of a drug and a device. Finally, the statute prohibits the sale
of any drug or device that the manufacturer hasn’t demonstrated to be “safe”
and “effective” for its intended use.

Until 1996, the FDA
disclaimed any authority to
regulate tobacco products
absent assertion of therapeutic
benefit by the manufacturer.
But in 1996, in light of new
evidence about the addictive
qualities of nicotine, the FDA
concluded that it had jurisdiction to
regulate tobacco products.
The tobacco manufacturers challenged the rule, contending that the FDA lacks
authority to regulate tobacco products. They pointed out that Congress had
passed six pieces of legislation in the preceding 30 years addressing the use
of tobacco, but never sought to ban the sale of cigarettes, to minors or others.
The case presented a Chevron question: The agency had interpreted the statute
to give it power to regulate tobacco products, but the plaintiffs argued that
the statute did not extend to such products. However, the Supreme Court
struck down the agency’s rule and concluded that the FDA lacked power to
regulate tobacco.
The court advanced two principal rationales for this conclusion. First, the
court reasoned that if tobacco was in fact a drug or device within the meaning
of the statute, then the FDA would have no choice but to ban it because it is
not safe and effective for any therapeutic use. But Congress had made clear
in several other statutes that specifically regulated tobacco products that
it did not want a complete ban on tobacco. Therefore, the court concluded
that Congress must have decided not to give the FDA authority over tobacco
products in the first place.
But by concluding that Congress didn’t intend to give the agency this authority,
the court avoided the second part of the Chevron test, which only comes
into play when the statute is ambiguous. When it is ambiguous, the court
has to defer to the agency. The agency likely would have won if it had come
to that.

Suggested Reading
<< Gailmard and Patty, Learning While
Governing.
<< Merrill, “The Story of Chevron.”
<< Merrill and Hickman, “Chevron’s Domain.”
<< Sunstein, “Interpreting Statutes in the

Regulatory State.”
<< ———, “Law and Administration after

Chevron.”
ÞÞ What role should an agency’s interpretation of the statute
that it is tasked with administering play in judicial
interpretation of the statute?
ÞÞ Does judicial deference to agency interpretations

represent a sensible accommodation of agency expertise,
or does it instead represent abdication of the courts’
constitutional role?

BIBLIOGRAPHY
Aberbach, Joel. Keeping a Watchful Eye: The Politics of Congressional Oversight.

Washington DC: Brookings Institution Press, 1990. This book is about
congressional control of agencies.
Achtenberg, David. “With Malice Towards Some: United States v. Kirby,

Malicious Prosecution, and the Fourteenth Amendment.” Rutgers Law Journal
(1995): 273–342. This article provides some interesting background about
United States v. Kirby, a case considered in Lecture 7.
Bean, Michael, and Melanie Rowland. The Evolution of National Wildlife Law.
Westport: Environmental Defense Fund and World Wildlife Fund, 1997. This
book provides an overview of environmental regulation to protect wildlife
and is a helpful background for Tennessee Valley Authority v. Hill, a case
considered in Lecture 6.
Calabresi, Guido. A Common Law for the Age of Statutes. Cambridge: Harvard
University Press, 1982. This book, by the former dean of Yale Law School and
a judge on the United States Court of Appeals, develops a pragmatic approach
to statutory interpretation.
Calabresi, Steven, and Kevin Rhodes. “The Structural Constitution: Unitary

Executive, Plural Judiciary.” Harvard Law Review 105 (1992): 1153–1216. This
article develops the unitary executive theory, which is considered in Lecture 10.
Chomsky, Carol. “Unlocking the Mysteries of Holy Trinity: Spirit, Letter,

and History in Statutory Interpretation.” Columbia Law Review 100 (2000):
901–956. This article is an in-depth look at the court’s decision in the Holy
Trinity case, which is considered in Lecture 4.
Dickerson, Reed. The Interpretation and Application of Statutes. New York:

Little Brown, 1975. This book provides an overview of theories of statutory
interpretation.
Bibliography 103
Dworkin, Ronald. Law’s Empire. Cambridge: Belknap Press, 1986. This
book, by a prominent scholar of jurisprudence, presents a moral theory of
the judge’s role.
Easterbrook, Frank. “Text, History, and Structure in Statutory Interpretation.”

Harvard Journal of Law and Public Policy 17 (1994): 61–70. This article, by a
judge on the United States Court of Appeals, defends the theory of statutory
interpretation known as textualism.
Elhauge, Einer. Statutory Default Rules. Cambridge: Harvard University Press,

2008. This book provides an in-depth look at the canons of interpretation.
Eskridge, William, Jr. Dynamic Statutory Interpretation. Cambridge: Harvard

University Press, 1994. This book develops a dynamic theory of statutory
interpretation.
———. “Public Values in Statutory Interpretation.” University of Pennsylvania

Law Review 137 (1989): 1007–1095. This article elaborates on the author’s
theory of dynamic statutory interpretation.
Gailmard, Sean, and John Patty. Learning While Governing: Expertise and
Accountability in the Executive Branch. Chicago: University of Chicago
Press, 2012. This book is about the tension between expertise and political
accountability in the administrative state.
Hamburger, Philip. Is Administrative Law Unlawful? Chicago: University of

Chicago Press, 2014. This book relies on history to suggest that administrative
policy making is unconstitutional.
Hart, Henry, Jr., and Albert Sacks. The Legal Process. St. Paul: Foundation
Press, 1994. This book provides the classic defense of an approach to statutory
interpretation that seeks legislative purpose.
Kagan, Elena. “Presidential Administration.” Harvard Law Review 114 (2001):

2245–2384. This article, by a future Supreme Court justice, explains and
defends independent agencies.

Landes, William & Posner, Richard, The Economic Structure of Tort Law.
Cambridge: Harvard University Press, 1987. This book develops an economic
theory of regulation by the common-law tort system.
Lessig, Lawrence, and Cass Sunstein. “The President and the Administration.”
Columbia Law Review 94 (1994): 1–119. This article focuses on the president’s
power to remove executive officials.
Macey, Jonathan, & Miller, Geoffrey, “The Canons of Statutory Construction

and Judicial Preferences.” Vanderbilt Law Review 45 (1992): 647–672. This
article is about the canons of interpretation.
Manning, John, “The Absurdity Doctrine.” Harvard Law Review 116 (2003):
2,387–2,486. This article, by the future dean of Harvard Law School, is about
the absurdity doctrine of statutory interpretation.
Mashaw, Jerry & Harfst, David, The Struggle for Auto Safety. Cambridge:
Harvard University Press, 1990. This book is an account of the long quest to
adopt and implement vehicle-safety standards.
McGarity, Thomas, “Some Thoughts on ‘Deossifying’ the Rulemaking

Process.” Duke Law Journal 41 (1992): 1,385–1,462. This article is about hard-
look review and its effects on agency decision making.
Merrill, Thomas. “The Story of Chevron: The Making of an Accidental

Landmark.” Administrative Law Review 66 (2014): 253–283. This article gives
some background about the Chevron case.
Merrill, Thomas, and Kristin Hickman. “Chevron’s Domain.” Georgetown

Law Journal 89 (2001): 833–921. This article provides an in-depth look at the
Chevron doctrine.
Nelson, Caleb. “What Is Textualism?” Virginia Law Review 91 (2005): 347–418.

This article explains the theory of textualism.
Oleszek, Walter. Congressional Procedures and the Policy Process. Los Angeles:
Sage, 2016. This book provides a detailed look at the legislative process.
Bibliography 105
Popkin, William. Statutes in Court: The History and Theory of Statutory
Interpretation. Durham: Duke University Press, 1999. This book provides an
overview of the various theories of statutory interpretation.
Posner, Richard. “Legal Formalism, Legal Realism, and the Interpretation of

Statutes and the Constitution.” Case Western Reserve Law Review 37 (1986):
176–217. This book, by a former law professor and United States Court of
Appeals judge, develops a pragmatic approach to statutory interpretation.
———. “Statutory Interpretation—in the Classroom and in the Courtroom.”

University of Chicago Law Review 50 (1983): 800–822. This article provides
a judge and former law professor’s reflections on the theory and practice of
statutory interpretation.
Radin, Max. “A Short Way with Statutes.” Harvard Law Review 56 (1942):
388–424. This article defends the approach of seeking legislative purpose in
statutory interpretation.
Scalia, Antonin. A Matter of Interpretation: Federal Courts and the Law.

Princeton: Princeton University Press, 1997. This book includes a series of
essays about statutory interpretation, including a famous one by the former
Supreme Court justice.
Scalia, Antonin, and Bryan Garner. Reading Law: The Interpretation of Legal
Texts. St. Paul: Thompson: West, 2012. This book, by a former Supreme
Court Justice and a scholar of legal language, provides a theory of statutory
interpretation.
Schuck, Peter, Foundations of Administrative Law. New York: Foundation

Press, 1994. This book provides a concise overview of administrative law.
Sharkey, Cathy. “State Farm ‘With Teeth’: Heightened Judicial Review in the
Absence of Executive Oversight.” New York University Law Review 89 (2014):
1589–1657. This article describes hard-look review, which is considered in
Lecture 11.

Siegel, Jonathan. “What Statutory Drafting Errors Teach Us About Statutory
Interpretation.” George Washington Law Review 69 (2001): 309–366. This
article is about how courts should approach apparent errors in statutes.
Sunstein, Cass. “Interpreting Statutes in the Regulatory State.” Harvard Law

Review 103 (1989): 405–508. This article is about how courts should interpret
statutes in light of agency interpretations.
———. “Law and Administration after Chevron.” Columbia Law Review 90

(1990): 2,071–2,120. This article is about the Chevron doctrine.
———. Valuing Life: Humanizing the Regulatory State. Chicago: University of

Chicago Press, 2014. This book is about cost-benefit analysis as a basis for
regulation.
Vermeule, Adrian. “Legislative History and the Limits of Judicial Competence:

The Untold Story of Holy Trinity Church.” Stanford Law Review 50 (1998):
1,833–1,896. This article is an in-depth look at the court’s decision in the Holy
Trinity case, which we consider in Lecture 4.
Whalen, Charles & Whalen, Barbara, The Longest Debate: A Legislative History
of the 1964 Civil Rights Act. Santa Ana: Seven Locks Press, 1985. This book
provides a detailed look at the path of the Civil Rights Act of 1964 through
a divided Congress.
Bibliography 107
IMAGE CREDITS
Page Credit
3 MicrovOne/iStock/Getty Images.
3 Sudowoodo/iStock/Getty Images.
4 Yale Center for British Art/Public domain.
6 Laetizia Haessig/EyeEm/Getty Images.
8 MicrovOne/iStock/Getty Images.
10 Tony Weller/Getty Images.
12 ©Depositphotos.com/bettorodrigues.
14 Leonid Eremeychuk/iStock/Getty Images.
15 tzahiV/iStock/Getty Images.
16 Chercvc/iStock/Getty Images.
19 Drnadig/E+/Getty Images.
21 Library of Congress/LC-USZ62-111165/Public domain.
23 PeopleImages/E+/Getty Images.
24 YakobchukOlena/iStock/Getty Images.
25 Library of Congress/LC-USZ62-60141/Public domain.
27 Mariia Nikolaeva/iStock/Getty Images.
29 vadimguzhva/iStock/Getty Images.
30 HEX/Getty Images.
32 isitsharp/iStock/Getty Images.
35 Kutaytanir/iStock/Getty Images.
36 FabioFilzi/E+/Getty Images.
39 Paul Bradbury/Getty Images.
43 Tennessee Valley Authority/Wikimedia Commons/Public domain.
46 U.S. Fish and Wildlife Service/Wikimedia commons/Public domain.
46 Steve Petteway, photographer, Supreme Court of the United States/Wikimedia
Commons/Public domain.
48 Steve Petteway, photographer, Supreme Court of the United States/Wikimedia
Commons/Public domain.
51 Roman Becker/EyeEm/Getty Images.
52 Library of Congress/LC-DIG-hec-16408/Public domain.
52 Ghrzuzudu/iStock/Getty Images.
53 AlexYustus/iStock/Getty Images.
54 Frank O’Dwyer/flickr/CC BY-SA 2.0.
58 Yulia Buzaeva/iStock/Getty Images.

60 Retinafunk/flickr/CC BY 2.0
61 Frank Easterbrook/Wikimedia Commons/Public domain.
62 Stas_V/iStock/Getty Images.
69 VanderWolf-Images/iStock/Getty Images.
71 Benoitb/E+/Getty Images.
74 Hybrid Images/Cultura RF/Getty Images.
78 Kyoshino/E+/Getty Images.
80 iShootPhotosLLC/E+/Getty Images.
82 LC-USZ62-117121/Public domain.
83 U.S. National Archives and Records Administration/Public domain.
86 OceanFishing/iStock/Getty Images.
87 IconicBestiary/iStock/Getty Images.
91 Toronto History/flickr/CC BY 2.0.
92 Deepblue4you/E+/Getty Images.
95 Jed Share/Kaoru Share/Getty Images.
96 IIP Photo Archive/flickr/Public domain.
99 marchmeena29/iStock/Getty Images.
101 tomch/E+/Getty Images.
Icons ihorzigor/iStock/Getty Images.
Image Credits 109

TTC - Law School For Everyone, Legislation and Regulation

Uploaded by

Copyright:

Available Formats

TTC - Law School For Everyone, Legislation and Regulation

Uploaded by

Document Information

Original Title

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

TTC - Law School For Everyone, Legislation and Regulation

Uploaded by

Copyright:

Available Formats

Topic Subtopic

Professor Peter J. Smith

THE GREAT COURSES

4840 Westfields Boulevard Suite 500

Copyright © The Teaching Company, 2019

Printed in the United States of America

Arthur Selwyn Miller

The George Washington

Before joining the faculty at GW Law, Professor Smith was an attorney at

II Law School for Everyone: Legislation and Regulation

2• Regulation by Statute and by Common Law . . . . . . . . . . . . . . . 10

3• Legislation and the Administrative State . . . . . . . . . . . . . . . . . . 19

4• Touchstones of Statutory Interpretation . . . . . . . . . . . . . . . . . . 29

5• The Letter versus the Spirit of the Law . . . . . . . . . . . . . . . . . . . . 35

6• When Is Statutory Meaning Plain? . . . . . . . . . . . . . . . . . . . . . . . . 42

7• Semantic and Substantive Interpretive Rules . . . . . . . . . . . . . . 50

8• How Do Courts Really Interpret Statutes? . . . . . . . . . . . . . . . . . 58

9• Federal Agencies as Regulatory Bodies . . . . . . . . . . . . . . . . . . . 68

10• Political Control of Agency Decision Making . . . . . . . . . . . . . . . 77

11• Judicial Review of Agency Rulings . . . . . . . . . . . . . . . . . . . . . . . 86

12• Weighing Agency Interpretations of Statutes . . . . . . . . . . . . . . 95

Table of Contents III

The Teaching Company expressly DISCLAIMS LIABILITY for any DIRECT,

IV Law School for Everyone: Legislation and Regulation

This course begins by considering a deceptively simple statute banning

2 Law School for Everyone: Legislation and Regulation

This is the fascinating world of legislation and regulation. To introduce you to it,

SETTING THE STAGE

Has your son actually broken the law? That depends

Interpreting statutes and regulations, and understanding the rules by which

4 Law School for Everyone: Legislation and Regulation

To return to your son’s predicament, it is pointless to fume about the citation

Because the rule in question is a town ordinance, another way to look at it is

These approaches highlight an important difference between courses that

However, statutes usually do not include reasoning—that is, they do not

Lecture 1–Making Sense of Legislation and Regulation 5

Another feature of interpreting legislation and regulation is that the ambiguity

A legislature could take care of this problem by creating a bright-line rule

6 Law School for Everyone: Legislation and Regulation

LETTER AND SPIRIT

The ordinance clearly prohibits sleeping in the station, with no listed

Lecture 1–Making Sense of Legislation and Regulation 7

This raises questions: Is it the judge’s role to interpret a statute to avoid

OTHER INTERPRETATIVE QUESTIONS

Returning to the no-vehicles-in-the-park example, assume that its meaning

To answer this question, we would not only need to define

But if the term can embrace unforeseen circumstances,

8 Law School for Everyone: Legislation and Regulation

To answer such questions, it is necessary to learn more about the role of

<< Popkin, Statutes in Court.

<< Scalia, A Matter of Interpretation.

ÞÞ Is legal meaning ever really plain?

Lecture 1–Making Sense of Legislation and Regulation 9

Environmental regulation is an interesting case study. For example, questions

CONTRACTS AND TORT LAW

Lecture 2–Regulation by Statute and by Common Law 11

12 Law School for Everyone: Legislation and Regulation

Lecture 2–Regulation by Statute and by Common Law 13

Despite its drawbacks, there are certain important advantages of regulation