Appellees Brief — United States v. American Library Assn., Inc.

Supreme Court brief2003

Ask Donna

What actually matters in this document.

Text

No. 02-361

lp £004

IN THE

Supreme Court of the United States

OCTOBER TERM, 2002

_— >

UNITED STATES OF AMERICA, et ai.,

Appellants,

AMERICAN LIBRARY ASSOCIATION, INC., ez ai.,

ON APPEAL FROM

Appellees.

THE UNITED STATES DISTRICT COURT FOR THE

EASTERN DISTRICT OF PENNSYLVANIA

BRIEF OF APPELLEES

MULTNOMAH COUNTY PUBLIC LIBRARY, et al.

Stefan Presser

ACLU of Pennsylvania

125 South Ninth Street

Suite 701

Philadelphia, Pennsylvania 19107

(215) 592-1513

David L. Sobel

Electronic Privacy

Information Center

1718 Connecticut Avenue, N.W.

Suite 200

Washington, D.C. 20009

(202) 483-1140

Christopher A. Hansen

(Counsel of Record)

Ann Beeson

Steven R. Shapiro

American Civil Liberties

Union Foundation

125 Broad Street

New York, New York 10004

(212) 549-2500

Charles S. Sims

PROSKAUER ROSE LLP

1585 Broadway

New York, New York 10036

(212) 969-3000

(counsel continued on inside front cover)

IBEST AVAILABLE COPY

ine

Lee Tien

Electronic Frontier Foundation

454 Shotwell Street

San Francisco, California 94110

(415) 436-9333

FOR PLAINTIFF MULTNOMAH COUNTY

PUBLIC LIBRARY:

Thomas Spoonsler, County Attorney

Multnomah County

501 N.E. Hawthorne Boulevard

Suite 500

Portland, Oregon 97214

(503) 988-3138

TABLE OF CONTENTS

Page

ne GED serepevncesccencersnnsssnessnecemsesesoces iv

ST ae l

COUNTERSTATEMENT OF THE CASE..................... 2

A. The Three-Judge Court’s Findings of Fact ...... 3

1. Public Libraries and Internet Access........... 3

2. Statutory Requirements...................:0:0000++ 6

3. Available Blocking Programs .................... 7

4. Blocking of Protected Speech.................... 8

5. The Inevitability of Overblocking.............. 11

6. The Inadequacy of CIPA’s Unblocking

ee 15

7. The Inability of Any Library To Comply

i a ee 17

8. Less Restrictive Alternatives...................... 19

B. The Three-Judge Court’s Legal Analysis ........ 21

SUMMARY OF ARGUMENT ....00.......::cccccccceseeeeeeeeeeeee 22

ITT ss iideecersta nian tenstpnnennsennseneennssnesnanensesessonsns 24

Page

I. The Three-Judge Court’s Injunction

Should Be Affirmed Because CIPA

Induces Public Libraries To Violate The

i a atalial 24

A. CIPA’s Content-Based Restrictions

On Speech Are Subject to Strict

ID cccnaneenentinennneniitiniaiemneannel -24

1. Like Traditional Public Forums,

Internet Access at Public Libraries

Promotes Core First Amendment

2. Strict Scrutiny Applies Because

The Government Seeks to

Selectively Exclude Protected

Speech From A Vast Democratic

Pe icrncennennnnnensenineeneiatemnannnns 27

B. CIPA’s Content-Based Restrictions

tf ee 29

The Mandated Use of Blocking

5 Programs Is Not Narrowly Tailored

To Prevent Access to Illegal

TD cccsemnsssnsercenssnemsnmessenneeenenn 29

1. Libraries Have Other Less

Restrictive Means To Help Patrons

Find What They Want And Avoid

Unwanted And Illegal Content.......... 33

C. The Disabling Provisions Fail to Cure

eo) 35

a <«

Page

D. The Government’s Arguments Ignore

The Record and Misconstrue the

I 39

II. CIPA Imposes A Prior Restraint On

ER I SEAS 42

III. CIPA Imposes An Unconstitutional

Condition On Funding Of Internet

Access at Public Libraries ......000000000000.cc0... 45

ean NETITTI cieccecineseitniininisinsilannantaeiestneteensipsintiibitamiiaesiina 49

ill

TABLE OF AUTHORITIES

Page

Cases

Arkansas Education Telev. Comm'n v. Forbes,

OL ee 27, 40, 41

Ashcroft v. Free Speech Coalition,

en ee 32

Bantam Books, Inc. v. Sullivan,

gS 23, 36, 44, 45

Blount v. Rizzi,

eT 23, 36, 43

Board of County Comm'rs v. Umbehr,

ee Ge incteertsncieniinsasittetteiensnnseentinemnenssammiees : wee 45

Board of Educ. v. Pico,

ee 25

Butler v. Michigan,

a 22, 33

City of Madison Joint School Dist. No. 8 v. Wisconsin

Employment Relations Comm'n,

og ee 28

Cornelius v. NAACP Legal Def. & Educ. Fund, Inc.,

473 U.S. 788 (1985)..........ccc00e0 seaiandanmsiaieaaitaianeianentl 27, 41

Denver Area Educ. Telecom. Consortium, Inc. v. FCC,

eee 36

Drive In Theatres, Inc. v. Huskey,

Re ce EG Be icccecccsctctcnsntiesnccnsenesszcsscctecseteel 44

FCC v. League of Women Voters,

er ea MIE ctictinnieicitiiabaiesiineaa 28, 46, 48

First Nat'l Bank of Boston v. Bellotti,

Say PU: UII ITEIPTTIII ascinbinnciiateeincsaiinceicheapicihaiiiiiinaahe 46

Forsyth County v. Nationalist Movement,

ee 45

Freedman v. Maryland,

| eee 42, 45

International Soc 'y for Krishna Consciousness, Inc. v. Lee,

IRIE RER EE Sa enE TERI OEE a. 27

Interstate Circuit v. Dallas,

Teen 44

Kreimer v. Bureau of Police,

we 24, 27

Lamont v. Postmaster General,

ERE ee ee 36

Legal Servs. Corp. v. Velazquez,

ee ee Ta rnseerrcseieeniiinsennenedininsentiniiapesinamnccotaniiiadiniinias 28, 47

Mainstream Loudoun v. Bd. of Trustees of the Loudoun

County Library,

24 F. Supp. 2d 552 (E.D. Va. 1998)... cece eens 27, 43, 47

Page

Minarcini v. Strongsville City School Dist.,

Pi Tf 4g.) Caso 24

Near v. Minnesota,

POU 43

Perry Educ. Ass'n v. Perry Local Educators’ Ass n.,

O10 8 27, 48

Perry v. Sindermann,

OO ee 45

Procunier v. Martinez,

GOB TEE. FE Ce ccecncnennensssecensceenensrresesssstmnemneciemnnened 46

Reno v. American Civil Liberties Union,

BE en passim

Rosenberger v. Rector and Visitors of the University of

Virginia,

a ee ccenirnteccnincreinniainciceniinnnisinnuiiiigs 21, 28

Rust v. Sullivan, 500 U.S. 173 (1991) ..........cccceeeeeeeeeeees 46, 48

Sable Communications of California, Inc. v. FCC,

og! ee 22, 23, 29

~perr ier v. City of Birmingham,

oo | 45

South Dakota v. Dole,

EF 22

Southeastern Promotions, Ltd. v. Conrad,

REE vnc en ae 28, 44

Sund v. City of Wichita Falls,

121 F. Supp. 2d 530 (N.D. Tex. 2000)...........cccccccceseecseeees 25

The Florida Star v. B.J.F.,

ee iain icecream 32

Turner Broad. Sys., Inc. v. FCC,

nani isi etd neha teerriaceteacsiinainaaielaaieil 32

United States v. Playboy Entertainment Group, Inc.,

ee icecarrtiinstsceinrscinceaimnensmieranatnciatsitieiinniined passim

Virginia v. American Booksellers Ass'n,

RN TE 46

Watchtower Bible & Tract Society of New York, Inc. v.

Village of Stratton,

SI iicciaethditesstienrasiciaenrasien etadiernrranrnaaaiaaaaiieni 37

Widmar v. Vincent,

| een NODE 27, 41

Statutes

ER 1,6, 7, 15

AF ies Ob I cctcerccteresecierenensnennntesnn 1, 7, 15, 35, 45, 48

Other Authorities

EERE EEE env an eC Ro 26

INTRODUCTION

This Court should affirm the unanimous decision of a

three-judge court striking down the Children’s Intemet

Protection Act (CIPA), 20 U.S.C. § 9134 (f); 47 U.S.C. §

254 (h)(6), the first statute in our nation’s history that

imposes federal speech restrictions on local libraries around

the country. Based on extensive findings of fact from an

eight-day tral, the lower court correctly held that CIPA

induces public libraries to violate the First Amendment when

they offer free Internet access to patrons. The statute

requires libraries to install blocking programs that inevitably

censor a substantial amount of protected speech for adults

and minors. Indeed no blocking program offers content

categories that are limited -- or even tied in any way -- to

CIPA’s legal definitions of obscenity, child pornography, or

material that is “harmful to minors.” In contrast to the

ineffective blocking programs mandated by CIPA, which are

both overly broad and ultimately ineffective, libraries around

the country have devised a number of less restrictive ways to

assist patrons who wish to avoid content they find offensive.

Under this Court’s well-established First Amendment rules,

CIPA fails the strict scrutiny required of content-based

speech restrictions.

Ignoring the extensive tral record, the government

argues that CIPA should be upheld under rational basis

review because it is consistent with the traditional discretion

of libraries to select books for their collections. Yet under

CIPA, the federal government, not a local librarian, dictates

what speech libraries nationwide must exclude from the

patrons. The federal government, not a local librarian,

requires library patrons and staff to use blocking programs

manufactured by private companies for profit. It does so

knowing that these companies: (1) use their own broad

content categories to block speech rather than the statute’s

legal definitions of unprotected expression; (2) block huge

amounts of text though the statute requires blocking of

images only; (3) refuse to disclose their secret block lists to

the federal government, local governments, or libraries that

must install them; and (4) make their decisions without any

input from or regard for the views of librarians or local

communities, much less prior judicial review. The

government also knows that the blocking programs, if

installed: (1) will block library patrons and staff from

accessing speech that does not meet the categories of the law

or the separate, broader categories defined by the private

companies; (2) will block speech that local libranans and

local communities have chosen to provide to patrons; (3) will

block library patrons and staff from reading constitutionally

protected speech literally millions of times; and (4) cannot

reliably block the speech defined by the statute.

Far from an exercise of local discretion, CIPA

substitutes the federal government’s judgment for that of

librarians. After closely considering local concerns, 93% of

all public libraries chose not to mandate the use of blocking

programs. The federal government now seeks to override

those decisions, and to install private software companies as

censors in our nation’s libraries. Because CIPA threatens to

distort the democratic, speech-enhancing qualities of both

public libraries and the Intermet, the three-judge court

correctly enjoined its enforcement.

COUNTERSTATEMENT OF THE CASE :

Two suits challenging the constitutionality of CIPA

were filed in the United States District Court for the Eastern

District of Pennsylvania, and were consolidated. The

plaintiffs in this case (hereinafter “Multnomah plaintiffs”)

include large urban libraries serving Portland, Oregon and

Santa Cruz, California; library systems serving rural and

suburban communities in south central Wisconsin and

Westchester County, New York; and _ sstate library

associations in Connecticut, Maine, and Wisconsin. The

Multnomah plaintiffs also include seven individuals who use

their local libranes for Internet access. For example,

plaintiff Emmalyn Rood used the Internet at her library in

her early teens to “research issues relating to her sexual

identity.” J.S. App. 22a. Finally, the Multnomah plaintiffs

include eight Web sites that were blocked by major blocking

programs even though they provided no information that was

illegal. Two of the Web sites are for political candidates.

AfraidtoAsk.com and Planned Parenthood provide medical

information about sex. PlanetOut provides information of

interest to gay and lesbian communities. /d. at 23a-24a.

Pursuant to the statute, a special three-judge court

was convened. After a period of discovery, the court held an

eight-day tnal at which it heard the testimony of twenty

witnesses and admitted hundreds of exhibits, including

depositions. /d. at 6a. On May 31, 2002, before the statute

would have required libraries to install blocking programs,

the three-judge court unanimously concluded that the statute

was unconstitutional and enjoined its application. The court

made “extensive findings of fact,” id. at 7a, that consume

almost one hundred pages in the Appendix to the

Jurisdictional Statement. The government does not argue

that any of these facts are clearly erroneous.

A. The Three-Judge Court’s Findings of Fact

1. Public Libraries and Internet Access

The three-judge court made a number of findings

about the nature of public libraries and their provision of

Internet access to the public. Specifically, the court found

that libraries share a common mission to provide patrons

with the widest possible range of information and ideas. J.S.

App. 33a. See also e.g. J.A. 35-36 (Morgan testimony); 79-

80 (Cooper testimony); 423-442 (Chelton Rebuttal Report).

That principle was endorsed by all of the librarians that

testified below, and is incorporated into the American

Library Association’s Bill of Rights and Freedom to Read

Statement. J.A. 358-50, 366-73 (PX 1, 9). In practice,

however, libraries cannot carry every book on their shelves

because of limited space. Faced with this physical

constraint, librarians apply professional standards to select

particular books, tapes, and other matenals for their

collections in order to meet the needs of their local

communities. J.S. App. 35a; J.A. 36 (Morgan testimony). '

In making these assessments, librarians try to provide library

patrons with the books they want to read. Librarians do not,

generally speaking, limit their library collections to books

that the staff considers “worthy” or “educational.” E.g. J.A.

79-80 (Cooper testimony); 423-442 (Chelton Rebuttal

Report). Libraries routinely provide access to purely

recreational material, including some that is sexually

explicit. J.S. App. 32a-33a; See also eg. J.A. 40-41

(Morgan testimony); 80, 84-85 (Cooper testimony); 428 (PX

83, Chelton Rebuttal Report).

Librarians also routinely provide patrons with access

to materials not in their collections “through the use of

bibliographic access tools and interlibrary loan programs.”

J.S. App. 34a; J.A. 42-43 (Morgan testimony); 83-84

(Cooper testimony). They do not apply selection criteria

when using these methods; instead, they provide the patron

with any resource they can obtain. J.A. 79-80 (Cooper

testimony). Some libraries allow patrons to obtain

interlibrary loan materials without going through a librarian.

J.A. 83 (Cooper testimony).

' For example, one of the branches of the Portland, Oregon library stocks

Russian language materials for its Russian émigré population. J.A. 81

(Cooper testimony).

All of the defendants’ experts agreed that, either

through direct selection or through techniques such as

interlibrary loan, libraries should provide patrons with any

information they seek so long as the information is legal.

J.A. 266-67 (Cronin testimony), 293 (Davis testimony).

Librarians are trained to use just about any means to assist a

patron in obtaining information he or she seeks. J.S. App.

33-34a. Increasingly, they turn to the Internet, which “vastly

expands the amount of information available to patrons of

public libraries.” J.S. App. 36a.

“The vast majority of public libraries offer Internet

access to their patrons.” /d.; Jt. Tnal Stip. para 263. There

is an enormous demand for the service. J.S. App. 36a.

“Public libraries play an important role in providing Internet

access to citizens who would not otherwise possess it.” Jd.

Indeed, libraries are the primary source of Internet access for

poor families. /d. at 36a-37a, 130a.

The court found that the provision of Internet access

at public libraries is notably different than the selection of

materials for physical collections. /d. at 120a-127a.

Through the Internet, libranans provide access to a vast

range of Internet content regardless of any librarian’s

judgment that patrons will want to read any particular page

or that any particular page meets professional selection

standards. Jd. at 124a. “[{W)]hen public libraries provide

their patrons with Internet access, they intentionally open

their doors to vast amounts of speech that clearly lacks

sufficient quality to ever be considered for the library’s print

collection.” J.S. App. 123a. Indeed, “almost none” of the

Web sites made available will have been reviewed by “either

the library’s collection development staff or even the

filtering companies.” Jd. at 125a, 137a (libraries do not

exercise “editorial discretion” on the Internet). Library

Internet access works so that “[a]ny member of the public

with Internet access could ... tonight jot down a few musings

on any subject under the sun, and tomorrow those musings

would become part of public libraries’ online offerings.” /d.

at 125a. To the extent that libraries find it necessary to place

limits on Internet access, the vast majority of libraries merely

ration limited resources and make no _ content-based

exclusions. /d. at 36a.

2. Statutory Requirements

The Children’s Internet Protection Act (CIPA)

applies to every local library in the country that receives

funds under two popular federal programs. /d. at 14a-16a.

Under CIPA, libraries*must install and operate “technology

protection measures” on every computer that provides

Internet access. 47 U.S.C. § 254(h)(6); 20 U.S.C. § 9134(f).

These measures must be functioning on computers used by

staff as well as patrons, adults as well as minors, and

computers privately funded as well as those subsidized by

the federal program. J.S. App. 18a. When the computer is

being used by an adult, the “technology protection measures”

must “protect against access” to “visual depictions” that are

obscene or child pornography. When the computer is being

used by a minor (under 17), it must additionally prevent

access to “visual depictions” that are “harmful to minors.” 47

U.S.C. § 254(h)(6); 20 U.S.C. § 9134(f); JS. App. 202a.

The statute applies to any library that receives federal

funds from either the “e-rate” program or as an LSTA grant.

47 U.S.C. § 254(h); 20 U.S.C. § 9134; J.S. App. 14a-16a.

More than fifty percent of public libraries participate in one

or both of these programs; there are over 16,000 libraries

nationwide. GX 37 at 4, Public Libraries and the Internet

2000: Summary Findings and Data Tables, Tr. 10/28/02 at

47. If the library is receiving e-rate funding (most libraries

do), a library administrator “may disable the technology

protection measure concerned, during use by an adult, to

enable access for bona fide research or other lawful

purpose.” 47 U.S.C. § 254(h)(6)(D); J.S. App. 168a n.33. If

the library receives no e-rate funding, but does receive LSTA

funds, it may disable the “technology protection measure”

for adults or minors “for bona fide research or other lawful

purpose.” 20 U.S.C. § 9134(f)(1) and (3).

3. Available Blocking Programs

Because Internet blocking programs are the only

“technology protection measures” currently available for

libraries to comply with CIPA, the three-judge court made

extensive findings about their operation and efficacy. J.S.

App. 48a-94a. Internet blocking programs, or filters, are

software products created and sold by private companies for

profit. These products categorize and then block speech on

the Internet. For example, one well-known product,

Websense, has created 30 categories ranging from “Abortion

Advocacy” to “Job Search” to “Tasteless” and “Adult.” Jd.

at 50-S5la. If a library utilizes the software, persons

attempting to access a site on the category list will be

blocked from doing so. Jd.

There is no technology in existence, or likely to come

into existence, that can “protect against access” to the

categories defined by the statute. /d. at 6a, lla, 68a; Jt.

Trial Stip. para 274, 291. All of the existing products use

categories much less precisely defined, and much broader,

than the statute. J.S. App. 49a-5la. None even purports to

apply the legal categories defined by the statute. Jt. Tnal

Stip. para 291. “[T}here is no judicial involvement in the

creation of the filtering software companies’ category

definitions.” J.S. App. S5la. All of the existing products

apply their categories uniformly without regard to local

values or community standards. /d.

Although the statute only requires blocking of “visual

depictions,” none of the available producis categorizes based

solely on visual depictions and none blocks visual depictions

without also blocking text. /d. at 56a, 93a. Neither judges

nor librarians are involved in the products’ decision to

categorize and block particular Web sites. /d. at 5la, 53a; Jt.

Tnal Stip. para 292. Web pages that are blocked by the

products are not notified. J.S. App. 53a.

The products consider their block site lists “to be

proprietary information, and hence . . . unavailable to

customers or the general public for review, so that public

libraries that select categories when implementing filtering

software do not really know what they are blocking.” JS.

App. 7a; Jt. Trial Stip. para 290. Neither the federal

government, which mandates its use, nor the librarians who

must install it, nor the patrons and staff who must live with

it, can obtain a list of the speech that has been blocked. J.S.

App. 52a-53a.

All of the parties agreed, and the court found, that all

of the available products substantially overblock, i.e., they

block sites that do not fit either the category definitions

established by the companies or the differently (and more

narrowly) defined statutory categories. /d. at 7a, 8a, lla,

12a, 48a-94a. J.A. 131-223 (Nunberg testimony), 295-357

(Edelman testimony), 467-474 (PX 109, Janes report); PX

110-119, 122-123, 128-135, 162A-K, 165-169 (screen shots,

lodged with the Court). All of the products also substantially

underblock, i.e., they fail to block sites that meet the

category or statutory definitions. J.S. App. 22a, 68a-94a.

4. Blocking of Protected Speech

As the three-judge court concluded, the software

mandated by CIPA “erroneously block[s] a huge amount of

speech that is protected by the First Amendment.” J.S. App.

91a; 12a. The court estimated the number of Web pages

erroneously blocked to be “[a]t least tens of thousands . . .

even when considered against the filtering companies’ own

category definitions.” /d. at 93a. The number that are

blocked and do not meet the statute’s categories, is “of

course much higher.” /d. The court found that it is difficult

to calculate overblocking rates accurately, though a number

of witnesses tried. Jd. at 68a-79a. The court ultimately

found that even defendants’ expert identified rates of

overblocking of “between nearly 6% and 15%.” Jd. at 49a.

These rates “greatly understate the actual rates of

overblocking that occurs” for a variety of reasons. /d. at 79a,

75a-76a. Because of the vast amount of content on the Web

and large number of library patrons around the country who

access the Internet daily, defendants’ expert admitted that

even using his estimate, library patrons would be wrongly

denied access to Web sites millions of times. Finnell

testimony 4/1/02 at 178-179.

Multnomah plaintiffs’ expert, Mr. Edelman, devised

and implemented a process to identify Web pages blocked by

four leading commercial blocking programs that do not fit

within those programs’ self-defined blocking categones. Mr.

Edelman first created a database of Web page addresses,

including substantial portions of site listings from the

popular Yahoo and Google Web directories. Each address

from the database was tested against each blocking program

to determine which programs blocked which pages. Most of

these Web pages were tested multiple times over a period of

several months. Edelman submitted a CD-ROM that

contained screen shots of over 6,000 sites, most of which the

court found were examples of overblocking. J.S. App. 79a-

86a; J.A. 467-474 (Janes expert report), The CD-ROM,

which has been lodged with the Court, also contains an index

to the sites, listing the site name and information about it

from the site itself and from the Google and Yahoo index

system, the dates on which it was blocked, the product that

blocked it, and the product category assigned it. The court

found that “many times the number of pages that Edelman

identified are erroneously blocked by one or more of the

filtering programs that he tested.” J.S. App. 85a-86a.

Largely on the basis of its own analysis of the

Edelman CD-ROM, the three-judge court identified a

number of specific examples of sites or pages that had been

overblocked, whether measured by the statutory categones

or the broader product categories. Jd. at 86a-89a. Each of

the plaintiff Web sites was wrongly blocked. J.A. 299

(Edelman testimony); Supp. Lodging 22 (PX 122-23,

Edelman CDs). Two of those sites, by Wayne L. Parker and

Jeffery Pollock, were created by political candidates to

promote their candidacy. J.S. App. 24a. Another site,

afraidtoask.com, is operated by a doctor and provides

information about medical issues that many people find

embarrassing. For example, afraidtoask.com has a section

discussing normal variations in penis size. Many of those

accessing afraidtoask.com do so through mechanisms that

ensure their anonymity. J.A. 237 (Bertman testimony).

Another of the sites, planetout.com, is a gay rights site that

was one of the sites plaintiff Emmalyn Rood tried to access

as she was doing research about her sexual identity. J.A.

223-32 (Rood testimony). In addition, the undisputed

evidence at trial showed the following examples of wrongly

blocked sites: The Web page of the Republican National

Committee, JX 5 at 59, Walker deposition, Tr. 4/3/02 at 90; a

Calgary Firefighters Museum site, Supp. Lodging 22, 108-

109 (PX 123, 168); a site teaching piano playing, Supp.

Lodging 22, 99 (PX 123, 165); the cover of Cosmopolitan

magazine, Supp. Lodging 81 (PX 162G); a juggling site,

Supp. Lodging 72 (PX 162B); a site about Navy units, Supp.

Lodging 19-21 (PX 119); a teen health site on sexually

transmitted disease, PX 160A, Tr. 3/28/02 at 217; an

abstinence site, Supp. Lodging 22 (PX 123), PX 160E, Tr.

3/28/02 at 217; and a scholarly discussion of biblical

attitudes towards homosexuality, PX 160D, Tr. 3/28/02 at

217. The Edelman CD-ROM, provides numerous other

examples. Supp. Lodging 22 (PX 122-23); See also J.A.

187-199 (Nunberg testimony); Plaintiffs’ Joint Proposed

Findings of Fact, 221-263.

Library experts testified that at least several hundred

of the examples of overblocking were sites librarians would

affirmatively select for the library’s collection if traditional

selection methods applied to the Internet. J.S. App. 82a n.

17. For example, many libraries affirmatively refer patrons

to the site of the Lewis Knighten Cancer Center, but it was

blocked. Lipow testimony, 4/2/02 at 153; Supp. Lodging 31

(PX 129). Another site blocked was alphasearch, a major

search engine to which many libraries link. Lipow

testimony, 4/2/02 at 154; Supp. Lodging 37 (PX 130). All of

these sites are collected, and screen shots provided, in

Appendix A on the Edelman CD-ROM.

5. The Inevitability of Overblocking

The three-judge court also concluded that all

blocking programs inevitably overblock and underblock

because of limitations inherent in the technology, and made

lengthy findings in support of this conclusion. J.S. App.

48a-94a; 150a-15la. It ultimately found that there is no

“technology protection measure” that will do what CIPA

requires without also blocking access to a vast amount of

speech that is constitutionally protected for both adults and

minors. /d. at 7a, 12a, 13a, 48a-94a. As an initial matter, all

of the products necessarily overblock because the categories

of speech they use are broader than the categories defined by

the statute, and because their blocking decisions are based on

text as well as image. /d. at 5la, 60a; 74a; compare JS.

App. 49a-5la with 71a-18a; Jt. Trial Stip. para 312; see also

J.A. 181 (Nunberg testimony). A number of other factors

inevitably cause overblocking.

First, categorizing the Web is an impossible task in

part due to “the Internet’s size, rate of growth, rate of

change, and architecture.” J.S. App. 68a, 54a. The majority

of Web pages cannot be found by the methods used by the

software companies. /d. at 28a-29a, 56a. Those pages are

accessible through a variety of means by library patrons. /7.;

J.A. 156, 176 (Nunberg testimony). Even the portions of the

Web that software companies can find, the “indexable Web,”

consists of at least 2 billion pages. J.S. App. 30a.

Approximately 1.5 million pages are added each day. J.S.

App. 30a; See also J.A. 168 (Nunberg testimony). The

content of particular Web sites is also highly dynamic, with

pages constantly being removed or changed. J.S. App. 29a-

30a. “Individual Web pages have an average life span of

approximately 90 days.” /d.

Second, to review all of the dynamic content on the

Web, blocking software companies have hired “between

eight_and a a few dozen employees.” /d. at 60a. Software

companies must use a combination of automated search

techniques and human review to categorize the vast content

on the Web, both of which are subject to serious error. J.A.

162-168 (Nunberg testimony). At least one of the products

allows computers to make decisions about sites to block

without human review. J.S. App. 60a. These sites are listed

solely on the basis of text and without reference to any visual

image. Jt. Trial Stip. para 309. The products very rarely

review a site again once they have initially categorized it,

though content changes constantly. J.S. App. 53a, 64a-65a;

Jt. Trial Stip. para 323. It is obvious that a comprehensive

review and evaluation of Web sites by these companies is

impossible. J.S. App. 68a, 54a.

Third, the products are unable to differentially block

access to video files, audio files, chat rooms, or discussion

groups. Regardless of whether a movie, for example,

contains one or a dozen sexually explicit scenes, the products

must either block or fail to block the entire movie. J.A. 335

(Edelman testimony).

Fourth, the products overblock because they

primarily block at the level of the root domain rather than at

the level of an individual Web page. J.S. App. 6la. To take

one example trom the record, at least one product blocks all

of the pages in the online magazine Salon (www.salon.com),

no matter how innocuous, because there are some pages in

Salon that they believe meet their criteria. /d. at 62a. There

may be hundreds or thousands of pages under a single root

URL. Jt. Trial Stip. para 243.

Fifth, the Internet’s addressing system causes serious

overblocking or underblocking. Every Web page can be

located by its Web address (a URL such as

Www .uscourts.gov) or its IP address (a series of numbers).

J.S. App. 26a-28a, 62a. Millions of Web pages with vastly

different content may share a single IP address. /d. If the

software companies fail to block by IP address, users could

bypass the blocking program by accessing forbidden sites

using the IP address. If the product blocks by IP address,

users will be unable to access any of the multiple sites at a

given IP address, no matter how harmless the content. /d.

All of the companies block by IP address as well as URL.

J.A. 321 (Edelman testimony). One company said that it

received fifty complaints a week about overblocking caused

by IP blocking. J.A. 199 (Nunberg testimony).

Sixth, overblocking is caused by so-called “loophole”

sites. J.S. App. 62a-64a. Loophole sites allow a user to

access a site using a Web address other than its primary

address. There are three major kinds of loophole sites:

caches (e.g., Google provides its own cache copy of any

Web page it indexes), translation sites (which provide a Web

page roughly translated into another language), and

anonymizers (which allow access to a Web page while

masking the user’s identity). Jd. To take one of these

examples: Google searches for Web sites. When it finds a

site, such as afraidtoask.com, it automatically makes a copy

of that site (in a “cache”) and gives it a name such as

“http://216.239.57.100/...www.afraidtoask.com...” The

entire content of afraidtoask can be accessed from this

alternate address as well as afraidtoask’s orginal address.

The blocking products must either prevent access to all

loophole sites, which “necessarily results in a significant

amount of overblocking, because the vast majority of pages

that are cached, for example, do not contain content that

would match a filtering company’s category definitions” or

must allow access “thus resulting in _ substantial

underblocking.” /d. at 64a. The Google cache alone

represents hundreds of millions of sites. J.A. 205 (Nunberg

testimony). All of the products block loophole sites. J.A.

200, 205 (Nunberg testimony).

Finally, the evidence established that all of the

blocking programs also inevitably underblock due to the

same inherent technological problems that lead to

overblocking. In addition, the companies’ human reviewers

only review a very small percentage of the Web. The vast

majority of Internet content is deemed not to meet the

blocking company categories and is thereby excluded from

human review by computers using an imprecise and secret

method of text analysis. J.S. App. 53a-64a. For that small

percentage not excluded by a computer, “[h]uman reviewers

generally focus on English language Web sites, and are

generally not required to be multi-lingual.” /d. at 60a. Even

with the products engaged, a library patron will be able to

find material that meets either the categories of the statute or

the categories set by the products. J.A. 179 (Nunberg

testimony); see also PX 121 at 27-8, Edelman expert report,

Tr. 4/2/02 at 127; JX 8 at 12, 164, Dussome deposition, Tr.

4/2/02 at 90; JX 9 at 93, Gallagher deposition, Tr. 4/2/02 at

90; JX 10 at 14, Blakeman deposition, Tr. 4/2/02 at 90.

me me

6. The Inadequacy of CIPA’s Unblocking

Provisions

The three-judge court also made findings about the

feasibility of unblocking sites wrongly blocked by the

programs, and the effect of requiring patrons to seek

permission to access blocked sites. CIPA allows, but does

not require, libraries to unblock sites upon the request of an

adult patron (or a minor in some cases) with a “bona fide

research or other lawful purpose.” 20 U.S.C. §9134(f)(3); 47

U.S.C. §254 (h) (6)(D). All of the blocking software

products offer a method for doing some unblocking. J.S.

App. 52a.

The feasibility of unblocking according to the

requirements of the statute was not established. The

products do not permit multiple people to unblock. Thus,

only one person in a library system, no matter how large or

busy the system, can have authority to unblock. J.A. 336-37,

339 (Edelman testimony). The products also do not permit a

librarian to unblock text while continuing to block only a

specific visual image or vice versa. J.A. 223 (Nunberg

testimony). The government failed to prove that any product

was capable of unblocking for adults but not minors, for one

patron or one terminal only (as opposed to all patrons or all

terminals), or for only a specified time period based on a

particular patron’s purpose. J.A. 341-342 (Edelman

testimony).

Even assuming that unblocking according to the

statute were feasible, the court found that “many patrons are

reluctant or unwilling to ask librarians to unblock Web pages

or sites that contain only materials that might be deemed

personal or embarrassing, even if they are not sexually

explicit or pomographic.” J.S. App. 47a, 172a. For

example, plaintiff Emmalyn Rood testified “that she would

have been unwilling as a young teen to ask a librarian to

disable filtering software so that she could view materials

concerning gay and lesbian issues.” J.S. App. 47a.

Similarly, plaintiff Mark Brown testified that he would have

been unwilling to ask a librarian to unblock a site when

doing research on his mother’s breast cancer. J.S. App.

172a-73a. Plaintiff Dr. Bertman testified that many users

came to his site, afraidtoask.com, through one of the

“loophole” methods, an anonymizer, to preserve their

anonymity. J.A. 237 (Bertman testimony). Thus, even

though he is a physician, and even though his site assures

readers that it will protect their privacy, his readers did not

even want him to know the address of the computer from

which they accessed his information.

The court also found that “[t]he pattern of patron

requests to unblock specific URLs in the various libranes

involved in this case” confirmed that “patrons are largely

unwilling to make unblocking requests unless they are

permitted to do so anonymously.” J.S. App. 47a. For

example, defendants’ expert testified that the Greenville

Public Library in South Carolina wrongly blocked close to a

hundred sites in a two-week period (a serious underestimate

of actual overblocking, as the court found), but the library

has received only twenty-eight unblocking requests in almost

two years. Id.

Even when sites were unblocked, the process took

“between 24 hours and a week.” /d. at 46a. Because most

libraries have more demand for Internet access than Internet

access terminals, they often ismit the time each patron may

use the computer. J.A. 73-74 (Morgan testimony).

Librarians testified that few patrons would interrupt their

limited access time to request an override. /d.; J.S. App.

174a.

eee

7. The Inability of Any Library To

Comply with the Statute

The government did not provide evidence of a single

library that was in compliance with the statute. The

government offered testimony from some libraries that

currently use blocking software. All but one refused to block

staff computers, although staff blocking is required by the

statute. Ewick testimony 4/3/02 at 31; J.A. 259 (Sudduth

testimony). These libraries made both blocking and

unblocking decisions based on the categories defined by the

software companies, not the more limited categories defined

by the statute. JX 1 at 42-43; Dooley Deposition, Tr. 4/3/02

at 90; JX 2 at 35-36, Majilton Deposition, Tr. 4/3/02 at 90;

JX 3 at 77, Stewart Deposition Tr. 4/3/02 at 90; JX 4 at 25-

26, Saferite Deposition Tr. 4/3/02 at 90; JX 5 at 32-34,

Walker Deposition, Tr. 4/3/02 at 90; Biek testimony 3/28/02

at 111-12; James testimony 3/29/02 at 15; Belk testimony

3/29/02 at 58. The Westerville, Ohio and Greenville, South

Carolina libraries blocked categories as broad as “tasteless.”

Barlow testimony 4/1/02 at 26; James testimony 3/29/02 at

16. Though the statute requires blocking only of images, all

but one library blocked text as well as images, and made

unblocking decisions based on text as well as images. E.g.

Barlow testimony 4/1/02 at 52-54; James testimony 3/29/02

at 15. Tacoma, which blocked only images, blocked any

image, no matter how innocuous, on a page that met the

blocking software company’s category. Biek testimony

3/28/02 at 48. “None of these libraries [proffered by the

government] makes differential unblocking decisions based

on the patrons’ age. Unblocking decisions are usually made

identically for adults and minors. Unblocking decisions even

for adults are usually based on suitability of the Web site for

minors.” J.S. App. 47a; JX 4 at 14, 45, 70, Saferite

deposition, Tr. 4/3/02 at 90; JX 5 at 34, Walker deposition,

Tr. 4/3/02 at 90; Biek testimony 3/28/02 at 113; James

testimony 3/29/02 at 15; Belk testimony 3/29/02 at 71;

Barlow testimony 4/1/02 at 42, 44. The libraries all blocked

or unblocked a site for all their patrons for all time; none

unblocked for only one patron or for a limited time period.

JX 1 at 48-49, Dooley deposition Tr. 4/3/02 at 90; Barlow

testimony 4/1/02 at 33; Ewick testimony 4/3/02 at 33, 57.

See also J.A. 341-42 (Edelman testimony). Though the

statute permits unblocking only for “bona fide research

purposes,” no library used this standard in determining

whether to unblock a site. E.g. Belk Testimony 3/29/02 at

71-73.

Government witnesses who currently use blocking

software in their libraries all admitted that the programs both

overblocked and underblocked. E.g. J.S. App. 77a-78a.

Indeed, even in Tacoma where a librarian claims to review a

list of every single site blocked to see if he can find errors,

there is still overblocking. Id. at 46a, 78a, 173a.

Defendants’ experts agreed that all of the libraries they

studied demonstrated both overblocking and underblocking.

Id. at 71a-79a. No library was able to unblock sites without

a delay, often of up to a week. Jd. at 174a.

The use of blocking software reached stark levels of

irrationality and arbitrariness. For example, in Tacoma, the

library on which the government most heavily relied, the

witness testified that the software blocked patrons from

viewing the pictures in the online version of Playboy

magazine. If asked to unblock these pictures, even by a

middle-aged, female sex researcher, he said he would refuse

to do so. At the same time, the library allowed all patrons,

regardless of age, to view the same pictures in Playboy

magazine on microfilm machines that were functionally

indistinguishable from the library’s computers. J.A. 250-252

(Biek testimony).

8. Less Restrictive Alternatives

Prior to the hammerlock imposed by CIPA, more

than 90% of public libraries had exercised their local

discretion not to require the use of blocking programs, in part

because of their deficiencies. J.S. App. 3a, 45a; Jt. Trial

Stip. para 271; see also J.A. 54 (Morgan testimony); 104

(Cooper testimony). Instead, as the three-judge court found,

libranes have developed a variety of methods for assisting

patrons in finding the content they want and avoiding

unwanted content. J.S. App. 37a-48a.

Some libraries offer patrons the option of voluntarily

using blocking software. Jd. at 45a; J.A. 53 (Morgan

testimony). Some of those libraries allow parents to decide

whether their child will use blocking software or not. J.A.

53 (Morgan testimony); J.S. App. 162a-163a. When library

patrons are given the chance to use blocking software, 80%

refuse to use the software. J.A. 56 (Morgan testimony).

Many libraries also offer “training to patrons on how

to use the Internet, including how to access the information

they want and to avoid the materials they do not want.” J.S.

App. 4la. Among other things, libraries offer links to sites

that have been screened by librarians and found to be

especially valuable for library patrons. /d. at 4la-42a. These

links do “not preclude patrons from attempting to access

other Internet Web sites.” /d. at 42a.

In addition, libraries have devised a variety of

techniques for minimizing the risk that a patron will view

material he or she finds objectionable on a computer being

used by another patron. /d. at 42a-44a. Some libraries place

their monitors so that it is difficult for non-users to see the

screen. Others offer “privacy screens” or provide recessed

monitors so that passers-by cannot see another user’s screen.

Id. Other libraries put the monitors in the middle of high

traffic areas to inhibit users from accessing objectionable

material. J.S. App. 44a. Libraries also offer software that

automatically erases a screen if there is no activity by a

patron for a specified time in order to assure that the next

user not be subjected to material that user finds

objectionable. J.A. 58 (Morgan testimony).

Though the government seeks to justify use of

blocking software by describing behavioral problems in the

library, the three-judge court found that behavioral problems

existed in libraries before there was an Internet and persist

even in libraries that use blocking software. Libraries

continue to deal with behavior problems by enforcing

general behavioral rules. J.S. App. 41a, 146a-47a; J.A. 462-

66 (PX 106, Multnomah County Library Policy).

Most libraries also have Internet use policies that

prohibit users from accessing illegal material. J.S. 37a. In

the event that librarians observe material on a user’s Internet

screen that is clearly illegal, they can and do call the police.

J.A. 61 (Morgan testimony); 453-61 (PX 105, Multnomah

County Library policy); see also Hamon testimony, 3/25/02

at 217-221.

Methods other than blocking software have proven

very effective in minimizing any problems caused by some

patrons objecting to material in the library. Libraries that did

not use mandatory blocking software testified that they had

very few complaints related to the content of Internet sites.

J.A. 61-69 (Morgan testimony); 109-111 (Cooper

testimony). The three-judge court found that the government

had not provided any evidence of the relative feasibility,

cost, and effectiveness of these various alternatives to

blocking software. J.S. App. 45a, 146a.

St ee

B. The Three-Judge Court’s Legal Analysis

The three-judge court held that CIPA was

unconstitutional because it induces libraries “to engage in

activities that would themselves be unconstitutional.” Jd. at

97a. The court analogized to the public forum doctrine,

finding that “[a]lthough a public library’s provision of

Internet access does not resemble the conventional notion of

a forum as a well-defined physical space, the same First

Amendment standards apply.” Id. at 108a (citing

Rosenberger v. Rector and Visitors of the University of

Virginia, 515 U.S. 819 (1995)). More specifically, the court

found that strict scrutiny applies because CIPA — and the

blocking programs it imposes - single out disfavored speech

for exclusion based on content in a forum otherwise

designated for unrestricted expressive activity on a wide

range of topics. /d. at 118a.

Although the court found that the state would have a

compelling interest in preventing access to illegal speech, the

government had to prove that CIPA’s blocking mandate “is

narrowly. tailored to further those interests, and that no less

restrictive means of promoting those interests exists.” /d at

148a. “Given the substantial amount of constitutionally

protected speech blocked by filters studied,” the court

concluded that CIPA was “not narrowly tailored.” J.S. App.

149a. The court also held that “there are plausible, less

restrictive alternatives to the use of software filters that

would serve the government's interest.” /d. at 158a. Finally,

the court concluded that the disabling provisions of CIPA did

not cure its unconstitutionality. “[{T]he content-based burden

that the library’s use of software filters places on patrons’

access to speech suffers from the same constitutional

deficiencies as a complete ban on patrons’ access to speech

that was erroneously blocked by filters, since patrons will

often be deterred from asking the library to unblock a site

and patron requests cannot be immediately reviewed.” /d. at

176a.

SUMMARY OF ARGUMENT

As the three-judge court recognized, “[t}he legal

context in which this extensive factual record is se: is

complex,” and “[{t}here are a number of potential entry points

into the analysis.” Jd. at 9a. Ultimately, under all of the core

constitutional principles, CIPA “lacks the precision that the

First Amendment requires when a statute regulates the

content of speech.” Reno v. American Civil Liberites Union,

521 U.S. 844, 874 (1997).

1. CIPA regulates protected speech on the basis of its

content. Under this Court’s longstanding precedents it would

clearly be unconstitutional if imposed as a direct mandate on

public libraries. See Reno, 521 U.S. at 879; United States v.

Playboy Entertainment Group, Inc., 529 U.S. 803, 813

(2000); Sable Communications of California, Inc. v. FCC,

492 U.S. 115, 126 (1989); Butler v. Michigan, 352 U.S. 380

(1957). CIPA’s censorship system is no less constitutional

simply because Congress chose to impose it indirectly on

libraries. As the government concedes, Congress may not

use its spending authority “to induce the States to engage in

activities that would themselves be unconstitutional.” South

Dakota v. Dole, 483 U.S. 203, 210 (1987). CIPA fails this

test because it induces libraries to use blocking software that

inevitably blocks access to a huge amount of protected

speech.

Internet access at public libraries uniquely promotes

First Amendment values. Because CIPA seeks to selectively

and arbitrarily exclude protected speech from a vast

democratic forum otherwise dedicated to unrestricted online

communication, it is properly subject to strict scrutiny.

Applying strict scrutiny, CIPA must be struck down because

the government failed to prove it is narrowly tailored to

serve a compelling government interest. See Reno, 521 US.

at 879. Playboy, 529 U.S. at 813; Sable, 492 U.S. at 126.

2. CIPA’s blocking mandate also imposes an

unlawful prior restraint by effectively silencing speech prior

to its dissemination in public libraries, without judicial

review or even the semblance of First Amendment due

process.

3. As a condition on the receipt of a federal

benefit, CIPA is equally unconstitutional. It requires

libranes to block speech even where Internet terminals are

wholly funded by non-federal dollars, and distorts the usual

function of public libraries in providing uncensored access to

information on the Internet.

In summary, under any legal theory, government-

mandated blocking programs are blunt instruments in an area

that requires far more sensitive tools. See Bantam Books,

Inc. v. Sullivan, 372 US. 58, 66 (1963). “The First

Amendment requires the precision of a scalpel, not a

sledgehammer.” J.S. App. 156a. As this Court has

explained, and as the case so clearly demonstrates, “the line

between speech unconditionally guaranteed and speech

which may legitimately be regulated,” suppressed, or

punished “is finely drawn.” Blount v. Rizzi, 400 U.S. 410,

417 (1971) (quoting Speiser v. Randall, 357 U.S. 513, 525

(1958)). Error in marking that line “exacts an extraordinary

cost.” Playboy, 529 US. at 817-18.

ARGUMENT

I. The Three-Judge Court’s Injunction Should Be

Affirmed Because CIPA Induces Public Libraries

To Violate The First Amendment.

A. CIPA’s Content-Based Restrictions On

Speech Are Subject to Strict Scrutiny.

1. Like Traditional Public Forums,

Internet Access at Public Libraries

Promotes Core First Amendment

Values.

If there was ever a context in which government

should not be allowed to impose content controls through its

spending power, it is Internet access at public libraries. As

the three-judge court correctly found, public libraries and the

Internet have many qualities analogous to traditional public

forums, which are entitled to the utmost protection of the

First Amendment. Sidewalks and public parks are protected

by strict scrutiny because their “speech-facilitating character

. . . makes them distinctly deserving of First Amendment

protection.” J.S. App. 129a. Like these traditional forums,

Internet access at public libraries “uniquely promote[s] First

Amendment values.” /d. The very purpose of public

libraries is to provide free access to books, ideas, resources,

and information for education, employment, enjoyment and

self-government. J.A. 35-36 (Morgan testimony), 79-80

(Cooper testimony); 423-442 (Chelton Rebuttal Report).

The public library is “the quintessential locus for the receipt

of information.” Kreimer v. Bureau of Police, 958 F.2d

1242, 1255 (3d Cir. 1992); see also Minarcini v. Strongsville

City School Dist., 541 F.2d 577, 582 (6™ Cir. 1976) (“A

library is a mighty resource in the free marketplace of

ideas.”’) (citing Abrams v. United States, 250 U.S. 616 (1928)

(Holmes, J., dissenting); Sund v. City of Wichita Falls, 121

F. Supp. 2d 530, 547 (N.D. Tex. 2000) (“The right to receive

information is vigorously enforced in the context of a public

library.”). Like other traditional forums, public libraries are

“generally open to any member of the public” who enters

them. J.S.App. 129a-130a. By providing free access to

information, public libraries have always played an

equalizing role that allows all individuals, regardless of their

income level, to participate fully in our democracy by

becoming informed, literate, educated, and culturally

enriched. Especially in public libraries, “the State may not,

consistently with the spirit of the First Amendment, ‘contract

the spectrum of available knowledge.’”” Board of Educ. v.

~Pico, 457 U.S. 853, 866 (1982) (quoting Griswold yv.

Connecticut, 381 U.S. 479, 482 (1965)).

The Internet has now provided unprecedented

opportunities to expand the scope of information from

around the globe that is available to users in public libraries, ~

at substantially less cost than comparable amounts of printed

material. Though the Internet “provides relatively unlimited,

low-cost capacity for communication of all kinds,’ Reno,

521 U.S. at 870, large segments of our society still lack

access to the Internet at home or work. J.S. App. 36a-37a.

“By providing Internet access to millions of Americans to

whom such access would otherwise be unavailable, public

libraries play a critical role in bridging the digital divide

separating those with access to new information technologies

from those that lack access.” Jd. at 130a and n.26.

In addition to opening their doors to all members of

the public seeking information, public libraries are like

traditional public forums because they provide broad access

to a wide range of speakers when they offer Internet access.

“By providing patrons with Internet access, public libranes

in effect open their doors to an unlimited number of potential

speakers around the world, inviting the speech of any

member of the public who wishes to communicate with

library patrons via the Internet.” J.S. App. 134a. As this

Court found, content on the Internet is “as diverse as human

thought.” Reno, 521 U.S. at 870.

In fact, the volume and diversity of speech on the

Internet “far exceeds the volume of speech available to

audiences in traditional public fora.” J.S. App. 134a. As this

Court explained in striking down prior attempts to censor

this important medium,

Through the use of chat rooms, any person

with a phone line can become a town crier

with a voice that resonates farther than it

could from any soapbox. Through the use of

Web pages, mail exploders, and newsgroups,

the same individual can become a

pamphleteer.

Reno, 521 U.S. at 870. See also LESSIG, CODE 167, 185

(1999) (“The architecture of the Internet, as it is ight now, is

perhaps the most important model of free speech since the

founding. . . . The model for speech that the framers

embraced was the model of the Internet — distribute, non-

centralized, fully free and diverse.’’).

Because of the Internet’s speech-enhancing qualities,

“the growth of the Internet has been and continues to be

phenomenal,” /d. at 885. “Technology expands the capacity

to cnoose; and it denies the potential of this revolution if we

assume the Government is best positioned to make these

choices for us.” Playboy, 529 U.S. at 818. Internet access at

public libraries “presents unique possibilities for promoting

First Amendment values,” J.S. App. 136a, and the three-

judge court thus correctly held that CIPA’s content-based

restrictions are subject to strict scrutiny. J.S. App. 138a.

2. Strict Scrutiny Applies Because the

Government Seeks to _ Selectively

Exclude Protected Speech From a Vast

Democratic Forum.

Library Internet access is also subject to strict

scrutiny when viewed as a designated public forum, because

the government seeks to selectively and arbitrarily exclude

protected speech from a forum otherwise dedicated to

unrestricted online communication. A designated public

forum is “property that the State has opened for expressive

activity by part or all of the public.” Jnternational Soc'y for

Krishna Consciousness, Inc. v. Lee, 505 U.S. 672, 678

(1992); Arkansas Education Telev. Comm'n v. Forbes, 523

U.S. 666, 678 (1998). Regulation that excludes disfavored

content from a designated forum is “subject to the same

limitations as that governing a traditional public forum.”

Lee, 505 U.S. at 678; see also Mainstream Loudoun v. Bd. of

Trustees of the Loudoun County Library, 24 F. Supp. 2d 552,

563 (E.D. Va. 1998); cf Kreimer, 958 F.2d at 1259. The

three-judge court correctly noted that “the relevant forum for

analysis is not the library’s entire collection, which includes

both print and ‘electronic media, . . . but rather the specific

forum created when the library provides its patrons with

Internet access.” J.S. App. 108a; see Cornelius v. NAACP

Legal Def. & Educ. Fund, Inc., 473 U.S. 788, 801 (1985);

Perry Educ. Ass'n v. Perry Local Educators’ Ass'‘n., 460

U.S. 37 (1983); Widmar v. Vincent, 454 U.S. 263 (1981).

Internet access at public libraries easily qualifies as a

designated public forum. As the lower court explained,

“where the state designates a forum for expressive activity

and opens the forum for speech by the public at large on a

wide range of topics, strict scrutiny applies to restrictions

that single out for exclusion from the forum particular speech

whose content is disfavored.” J.S. App. 118a. Conversely,

“the more narrow the range of speech that the government

chooses to subsidize (whether directly, through government

grants or other funding, or indirectly, through the creation of

a public forum) the more deference the First Amendment

accords the government in drawing’ content-based

distinctions.” Jd. at 112a. Thus, in Southeastern

Promotions, Ltd. v. Conrad, 420 U.S. 546, 559 (1975), this

Court applied heightened scrutiny to invalidate a town’s

exclusion of the rock musical “Hair” from a theater

otherwise open for the public at large to use for

performances. In Rosenberger, 515 U.S. at 833, this Court

applied strict scrutiny in stniking down funding restrictions

on student newspapers because the funding program was

designed to “encourage a diversity of views from private

speakers.” And in FCC v. League of Women Voters, 468

U.S. 364 (1984), this Court invalidated prohibitions against

editorializing by public radio networks under strict scrutiny

because the federal program broadly funded diverse speech

on a wide range of subjects. See also Legal Servs. Corp. v.

Velazquez, 531 U.S. 533, 543 (2001); City of Madison Joint

School Dist. No. 8 v. Wisconsin Employment Relations

Comm'n, 429 U.S. 167 (1976) (holding that school board

may not exciude certain speakers from meetings open to

direct citizen involvement).

Applying these principles to this case, there may be

no better example of a diverse range of speakers, viewpoints,

and readers than Internet access at public libraries. “The

unique speech-enhancing character of Internet use in public

libraries derives from the openness of the public library to

any member of the public seeking to receive information,

and the openness of the Internet to any member of the public

who wishes to speak.” J.S. App. 135a-136a. “[B]y

providing patrons with even filtered access, the library

permits patrons to receive speech on a virtually unlimited

number of topics, from a virtually unlimited number of

speakers, without attempting to restrict patrons’ access to

speech that the library, in the exercise of its professional

-- n

judgment, determines to be particularly valuable.” Jd. at

121a.

When public libraries provide Internet access, they

“create[] a forum for the facilitation of speech, almost none

of which either the library’s collection development staff or

even the filtering companies have ever reviewed.” Jd. at

125a. By forcing libranes to use blocking programs, CIPA

“risk[{s] fundamentally distorting the unique marketplace of

ideas that public libraries create when they open their

collections, via the Internet, to the speech of millions of

individuals around the world on a virtually limitless number

of subjects.” /d. at 126a.

Because CIPA is properly subject to strict scrutiny, it

is presumptively invalid, and must be struck down unless the

government can prove it is narrowly tailored to serve a

compelling government interest. See Reno, 521 U.S. at 879;

Playboy, 529 U.S. at 813; Sable, 492 U.S. at 126. Over

ninety pages of factual findings, none disputed by the

government, establish that the government failed to meet

their burden in this case.

B. CIPA’s Content-Based Restrictions Fail Strict

Scrutiny.

1. The Mandated Use of Blocking

Programs Is Not Narrowly Tailored To

Prevent Access to Illegal Speech.

As the three-judge court’s detailed findings clearly

establish, CIPA suppresses a vast amount of Internet content,

that is fully protected by the First Amendment, and is thus

not narrowly tailored to serve the government’s interest in

prohibiting access to illegal images. Blocking programs

“block many thousands of Web pages that are clearly not

harmful to minors, and many thousands more pages that,

while possibly harmful to minors, are neither obscene nor

child pornography.” J.S. App. 148a-149a. Even the

defendants’ own expert concluded that “between 6% and

15%” of blocked Web pages contain no content that meets

“even the filtering products’ own definitions of sexually

explicit content, let alone the legal definitions of obscenity or

child pornography.” Jd at 149a. This evidence “significantly

underestimate[s] the amount of speech that filters

erroneously block, and at best provide[s] a rough lower

bound on_ the filters’ rates of overblocking.” Id.

Defendant’s expert conceded that even using his estimate,

library patrons would be wrongly denied access to Web sites

millions of times. Finnell testimony 4/1/02 at 175-179.

Indeed, given the fundamental flaws of blocking

programs, the three judge court found that the mandatory use

of such programs by public libraries is inherently too blunt a

tool to satisfy the constitutional requirements of narrow

tailoring. J.S. App. 152a. “[A]ny technology protection

measure that blocks a sufficient amount of speech to comply

with CIPA ... will necessarily block substantial amounts of

speech that does not fall within these categories.” Jd. at

15la. Because automated methods of reviewing content are

severely flawed,

[fliltering companies are left with the

Sisyphean task of using human review to

? The government’s argument that plaintiffs’ expert testimony established

only a 1% overblocking rate is a shocking mischaracterization of the

evidence and the court’s findings. Govt. Br. at 34-35. Edelman

specifically testified that his research did not attempt to establish an

overblocking percentage, due to the inherent difficulties of doing so. J.A.

296-306, 350-356 (Edelman testimony). The court itself issued detailed

factual findings about those difficulties, JS. App. 85a (accepting

government argument that Edelman’s results cannot be generalized

beyond his 6,775 sites), and found that all of the evidence significantly

underestimated the actual amount of overblocking. /d. at 85a-86a.

identify from among the approximately two

billion Web pages that exist, the 1.5 million

new pages that are created daily, and the

many thousands of pages whose content

changes from day to day, those particular

Web pages to be blocked.

Id. at 150a. To cope with the Web’s size, blocking programs

use a variety of techniques, which necessarily cause them to

block protected content. See also Section AS, supra.

They routinely block every page of a Web site that

contains only some content falling with the companies’ own

definitions; they block text as well as images; they block

entire Web sites containing fully protected speech simply

because those sites share IP addresses with sites that fall

within their category definitions; they block access to entire

video files, audio files, chat rooms, and discussion groups

regardless of how little content meets their category

definitions; and they generally fail to review pages again

once categorized to determine whether the content has

changed. J.S. App. 150a.

“(T]he filters required by CIPA block substantial

numbers of Web sites that even the most puritanical public

library patron would not find offensive.” J.S. App. 155a.

For example, the products blocked the Lakewood High

School alumni page and a political party in Uganda. J.A.

318; 326 PX 166, 169 (screen shot). They also blocked a

soccer site in the United Kingdom and The Journal of

Contemporary Obituaries. PX 162 I, 162 K (screen shots,

lodged) Finnell testimony 4/1/02 at 161-169.

The government failed to produce evidence of any

blocking program “that avoids overblocking a substantial

amount of protected speech.” Jd. at 15la. Yet it contends

that blocking millions of Web pages, most with content

nowhere close to the line of illegal speech, is a “negligible”

side effect of its mission to prevent access to illegal speech.

As this Court said just last term, “(t]he argument. . . that

protected speech may be banned as a means to ban

unprotected speech . . . turns the First Amendment upside

down. The Government may not suppress lawful speech as

the means to suppress unlawful speech.” Ashcroft v. Free

Speech Coalition, 535 U.S. 234, 122 S. Ct. 1389, 1401

(2002).

Just as all programs necessarily overblock, they all

necessarily fail to block access to sites that clearly fall within

their own category definitions for sexually explicit material,

and that may be illegal. J.S. App. 65a-66a. Because of the

nature of all classification systems, “[t]here is an inherent

tradeoff between any filter’s rate of overblocking . . . and its

rate of underblocking.” /d. at 66a. The government’s library

witnesses who used blocking programs all experienced

underblocking problems in their libraries. /d. at 77a. Even

with blocking programs installed, then, any determined

patron will easily be able to access sexually explicit material

on library computers. J.A. 179 (Nunberg testimony). CIPA

fails to achieve the government’s goal to prevent access to

illegal material “in a direct and material way.” Turner

Broad. Sys., Inc. v. FCC, 512 U.S. 622, 624 (1994). As

Justice Scalia wrote in The Florida Star v. B.J.F., “a law

cannot be regarded as . . . justifying a restriction upon

truthful speech, when it leaves appreciable damage to

[defendant’s] supposedly vital interest unprohibited.” 491

U.S. 524, 541-42 (1989) (Scalia, J., concurring). CIPA

clearly fails the strict scrutiny required when fundamental

First Amendment rights are at stake. Reno, 521 U.S. at 874.

2. Libraries Have Other Less Restrictive

Means to Help Patrons Find What

They Want and Avoid Unwanted and

Illegal Content.

Even assuming that the breadth of CIPA’s impact on

protected speech could ever be justified, see Butler v.

Michigan, 352 U.S. 380 (1957), the government clearly

failed to meet its “heavy burden ... to explain why a less

restrictive provision would not be as effective as [CIPA].”

Id. at 879. As the three-judge court found, a number of

alternative methods further the government’s stated interests

in a manner far less burdensome on protected speech than

the mandatory use of blocking programs for all adults and all

minors, including library staff. J.S. App. 157a-167a; see

generally Section A8, supra. The vast majority of America’s

libraries use these methods, and not mandatory blocking for

adults, to help patrons find information they want and avoid

unwanted and illegal content. Plaintiffs’ libraries testified

that they receive few complaints when using these methods.

Id. While these alternatives may not be perfect, the

government failed to prove that they are less effective than

blocking programs. As the lower court found, the

government offered no data “comparing the use of library

computers to access child pornography and . . . obscenity in

libraries that use blocking software and in libraries that use

alternative methods.” Jd. at 16la. Especially because

blocking programs are themselves far from effective, “i]t is

no response that [a less restrictive alternative] . . . may not go

perfectly every time.” Playboy, 529 U.S. at 824.

In particular, almost all libraries have Internet use

policies that forbid access to illegal content.

Libranes can ensure that their patrons are

aware of such policies by posting them in

prominent places in the library, requiring

patrons to sign forms agreeing to comply with

the policy before the library issues library

cards to patrons, and by presenting patrons,

when they log on to one of the library’s

Internet terminals, with a screen that requires

the user to agree to comply with the library’s

policy before allowing the user access to the

Internet.

J.S. App. 158a. Almost all libraries also offer training in

Internet usage, and provide lists of Web sites selected by

librarians as appropriate for children or of particular

relevance to their local communities. Section A8, supra.

Some libraries place unblocked computers in well-trafficked

areas to discourage violations of Internet use policies. Other

libraries have installed privacy screens or recessed monitors,

or have segregated computer terminals, to prevent unwanted

exposure to material viewed by others. J.S. App. 165a-166a.

Similar alternatives exist to prevent minors from

accessing material harmful to minors. Some libraries also

offer the optional use of blocking software. Others have

policies that allow parents to decide whether their children

will use blocking programs, or policies that require only

younger children to use them. /d. at 162a-163a. CIPA

replaces parental decision-making about appropriate Internet

use with the federal government’s blanket judgment. “A

court should not assume that a plausible, less restrictive

alternative would be ineffective; and a court should not

presume parents, given full information, will fail to act.”

Playboy, 529 U.S. at 824. In sum, the government “failed to

show that the less restrictive alternatives . . . are ineffective

at furthering the government’s interest.” J.S. App. 167a.

¢. The Disabling Provisions Fail to Cure

CIPA’s Defects.

The three-judge court held that CIPA’s disabling

provisions do not cure its constitutional defects. J.S. App.

167a-177a. For a host of reasons, that conclusion was

correct. As an initial matter, libraries that receive e-rate

funding may only override the software for adults with

“bona fide research or other lawful purpose[s].” 47 U.S.C. §

254(h)(6)(D). For teenagers in thousands of libraries across

the country, the provision is no help at all in alleviating

CIPA’s tremendous overblocking of protected speech.

Furthermore, any inquiry into a patron’s purpose is contrary

to good librarianship. J.A. 122 (Cooper testimony).

The provisions are also unworkable for practical

reasons. It is not possible to unblock visual images without

also unblocking text and vice versa. Section A6, supra. The

government also failed to prove that any software permits

unblocking for a single patron based on his purpose, while

continue to block access by those with less “bona fide”

purposes. /d. In addition, processing unblocking requests

could take up to a week. J.S. App. 46a.

To make matters worse, the unblocking provisions

are impossibly vague. J.S. App. 167a-170a. The statute

refers to the patron’s “purpose” and not the nature of the site.

Thus, it appears to require librarians to inquire into the “bona

fide” nature or “lawfulness” of the patron’s purpose in

seeking to access a site regardless of whether the site is

illegal. The government asks the Court to ignore the word

“purpose” and to treat the provision as permitting unblocking

of any lawful site. But this interpretation would render the

“bona fide research” language superfluous.

Even assuming the government’s interpretation

would provide any useful guidance to librarians, it would

place librarians in the business of making essentially

unreviewable decisions concerning the legal categories of

obscenity. This Court has repeatedly required that those

decisions be made by courts, not by administrative officials,

even in the context of non-criminal proceedings. Blount v.

Rizzi, 400 U.S. 410 (1971); Bantam Books v. Sullivan, 372

U.S. 58 (1963). Librarians have repeatedly disavowed any

expertise in making these legal judgments and expressed

certainty that the provision would result in arbitrary and

inconsistent judgments by different individual librarians with

different values and concerns. J.A. 124 (Cooper testimony).

Finally, even putting aside the provision’s

infeasibility, vagueness, and delegation of judgments to

administrative officials, the disabling provision still

unconstitutionally deters patrons from accessing protected

speech. As the lower court held, the “requirement that

library patrons ask a state actor’s permission to access

disfavored content violates the First Amendment.” J.S. App.

170a. This Court has struck down numerous content-based

restrictions that require recipients to identify themselves

before being granted the nght to access or communicate

disfavored speech. See, e.g., Lamont v. Postmaster General,

381 U.S. 301 (1965) (invalidating federal statute requiring

postmaster to halt delivery of communist propaganda absent

affirmative request); Denver Area Educ. Telecom.

Consortium, Inc. v. FCC, 518 U.S. 727 (1996) (striking

down federal law requiring cable users to request sexually

explicit programming in writing); Playboy, 529 U.S. at 803

(invalidating law requiring cable users to request access to

scrambled sexually explicit programming). As this Court

explained just last term, “It is offensive -- not only to the

values protected by the First Amendment, but to the very

notion of a free society -- that in the context of everyday

public discourse a citizen must first inform the government

of her desire to speak to her neighbors and then obtain a

permit to do so.” Watchtower Bible & Tract Society of New

York, Inc. v. Village of Stratton, 536 U.S. 150, 122 S.Ct.

2080, 2089 (2002) (striking down local ordinance that

prohibited door-to-door canvassers from “promoting any

cause” without first obtaining a permit). It is equally

offensive that a citizen must obtain government approval

before accessing protected but disfavored speech on the

Internet in her public library.

The deterrent effect of the disabling provisions is “a

matter of common sense as well as amply borne out by the

trial record.” J.S. App. 172a. For example, plaintiff

Emmalyn Rood testified that as a gay teen she would have

been unwilling to ask a librarian to disable blocking

programs so that she could research issues related to her

sexual identity. J.S. App. 47a. Plaintiff Mark Brown would

have been equally embarrassed to ask a librarian to unblock

sites when he was researching his mother’s breast cancer.

Id. at 172a-73a. Many patrons would also not want to

relinquish precious time during their Internet sessions to

request unblocking that may take up to week. J.A. 127

(Cooper testimony); J.S. App. 46a. Significantly, the three-

judge court found that the reluctance of patrons to request

unblocking is also established “by the low number of patron

unblocking requests, relative to the number of erroneously

blocked Web sites, in those public libraries that use software

filters and permit patrons to request access to incorrectly

blocked Web sites.” /d. at 173a.

Ignoring specific findings of the three-judge court,

the government argues that requiring patrons to ask

permission to view blocked sites is no different than other

questions patrons ask of librarians daily. The analogy misses

the mark. Librarians know from experience that patrons are

unwilling to discuss their private concerns, and for that

reason offer services to allow patrons to ask questions of

librarians, or check out books, even through interlibrary loan,

anonymously. J.A. 83 (Cooper testimony). Library patrons

can generally access print material without identifying

themselves, by simply taking a book off a shelf and reading -

it at a library desk.

Even more fundamentally, asking a librarian to

unblock a site that may be illegal is vastly different than

asking for help to find a book. When a book is in the

library’s catalogue, the library has sent the unmistakable

signal to the patron that it will assist the patron in getting that

book. When a patron attempts to access an Internet site and

is presented with a message, “THIS SITE IS BLOCKED BY

THE LIBRARY DUE TO FEDERAL LAW REQUIRING

BLOCKING OF OBSCENITY AND OTHER ILLEGAL

SPEECH,” the patron is sent an equally unmistakable signal

that the library disapproves of the site. The patron may not

know that the decision to block the site was not made by a

librarian, but by an employee of a software company in

California. The patron may also not know that the software

is astonishingly inaccurate. For example, if the software

blocks the Republican National Committee Web site, the

patron may wrougly conclude that the site is mis-named and

is really a site with sexually explicit material. The deterrent

effect of the disabling provisions thus far exceeds even the

common reluctance of patrons to seek assistance.

Given the content-based burden the disabling

provision imposes on protected speech, and its strong

deterrent effect, the provision “fail[s] to cure CIPA’s lack of

narrow tailoring.” Jd. at 177a; see also Playboy, 529 U.S. at

812 (“It is of no moment that the statute does not impose a

complete prohibition. The distinction between laws

burdening and laws banning speech is but a matter of

degree”’).

D. The Government’s Arguments Ignore the

Record and Misconstrue the Relevant Law.

Rather than grapple with the findings and

overwhelming evidence against them, the government argues

that the First Amendment is practically irrelevant to this case

because CIPA’s federal filtering requirement is consistent

with a library’s traditional collection decisions and thus

subject only to rational basis review. This analysis ignores

the three-judge court’s specific findings to the contrary, and

is flawed in numerous ways. First, CIPA imposes the federal

government’s censorship standard on libranes nationwide,

and is thus a far cry from a library’s exercise of its own

editorial judgment. Indeed, 93% of libraries had already

decided, after much review, not to mandate filters for all

users. CIPA directly overrides those local decisions.

Mandated filtering under CIPA, by definition, eliminates

rather than preserves library discretion.”

Second, the lower court found that there are

“fundamental differences” between the book selection

process and the provision of Internet access. J.S. App. 124a.

Unlike decisions to include books in their print collections,

when offering Internet access librarians do not exercise

editorial discretion and select only pre-approved speech for

inclusion. When public libraries provide patrons with

Internet access, they allow any member of the public to

> The government also implies that under the three-judge court's

rationale, there would be a “perverse incentive” for public libraries to

make available less information to preserve their discretion. The

government offered no evidence at trial to support the theory that

libraries would offer access only to preselected sites if they could not

mandate filters. Their own witnesses specifically rejected that approach.

J.S. App. 42a.

receive speech “from anyone around the world who wishes

to disseminate information over the Internet.” /d. at 137a-

138a. CIPA is thus easily distinguishable from cases cited

by the government which upheld the government’s exercise

of editorial discretion in selecting certain speech for

inclusion in a state-created forum. As the three-judge court

explained, in those cases “the state actor exercising the

editorial discretion has at least reviewed the content of the

speech that the forum facilitates.” J.S. App. 12la (citing

NEA vy. Finley, 524 U.S. 569 (1998) (NEA examined the

content of artworks it chose to subsidize on the basis of

artistic excellence); Forbes, 523 U.S. at 673 (public

broadcaster specifically reviewed and approved each speaker

permitted to participate in debate). In providing patrons

with even filtered Internet access, “a public library creates a

forum for the facilitation of speech, almost none of which

either the library’s collection development staff or even the

filtering companies have ever reviewed.” /d. at 125a, 123a,

7a. Unlike a public library’s decision to carry books on a

selected topic, “members of the general public . . . define the

content that public libraries make available to their patrons

on the Internet.” /d. at 124a-125a. Moreover each of the

cases cited by the government involved practical limits on

the number of speakers that could be accommodated; these

limits simply do not apply to Internet access.

Third, the government again misstates the record

when it suggests that using filters is akin to the reliance by

some libraries of third-party vendors or book reviews to

make book selection decisions. Govt. Br. at 27. These

resources are a transparent means of assisting librarians in

including material in their collections. As the three-judge

court specifically found, when using these resources

“librarians still retain ultimate control over their collection

development and review all of the materials that enter their

library’s collection.” J.S. App. 35a; see also /d. at 123a. In

stark contrast, blocking programs censor rather than select,

~ mo:

and libraries have no idea what Web sites the program

ultimately blocks. /d. at 125a.

Fourth, the government argues that there is no tenable

ground for subjecting book collection to rational basis

review and Internet access to strict scrutiny. Yet this Court

has recognized that there may well be different types of

forums, subject to differing levels of scrutiny, within the

same physical space or organizational structure. So, for

example, a televised debate with “an open-microphone

format” is subject to strict scrutiny, while a debate in which

the same station controls the content is not. Forbes, 523

U.S. at 680; cf Cornelius, 473 U.S. at 801-02 (recognizing

that different standards would apply to the federal charity

drive than to the federal workplace generally); Widmar, 454

U.S. at 264 (applying different standard to university

meeting facilities than to state university generally). In

addition, the issues that warrant less-than-strict-scrutiny in

collection decisions (the necessity of choice, in light of cost

and space constraints), are simply not present when Congress

mandates censorship at public library Internet terminals

nationwide.

Fifth, the government argues that the three-judge

court’s approach must be wrong because it “would nsk

transforming the role of public libraries in our society.”

Govt. Br. at 19. That argument is exactly backward. In fact,

it is CIPA that risks changing librarians from information

providers into censors. CIPA forces libraries to install filters

that, out of the “vasi democratic forum” of the Internet,

“single out for exclusion particular speech on the basis of its

disfavored content.” /d. at 138a.

Finally, the government suggests that mandated

filtering is reasonable because “a library patron would only

infrequently need to have access to a Web site that has been

blocked in order to obtain the information he or she seeks.”

Govt. Br. at 34. This argument approaches the absurd. It

implies that the Internet contains only facts, and not fiction,

art, opinion, and vast amounts of unique expression. By

analogy, the government would hardly sanction the removal

of Edgar Allen Poe from library shelves because Charles

Dickens remains available, or suggest the painter Yves

Tanguy as an appropriate substitute for Frida Kahlo, or the

Beatles for Beethoven. Even Web sites containing factual

information can, of course, be unique. Plaintiff Jeffrey

Pollock used his Web site to promote his campaign for

Congress; when it was blocked, it was little comfort that

Web users could reach his opponent’s site. This Court has

held that “one is not to have the exercise of his liberty of

expression in appropnate places abridged on the plea that it

may be exercised in some other place.” Reno, 521 U.S. at

880. The government suggests an even more twisted re-

writing of the First Amendment: that one speaker may be

silenced so long as another remains free to speak.*

II. CIPA Imposes A Prior Restraint On Speech.

CIPA’s blocking mandate also imposes an unlawful

prior restraint by effectively silencing speech prior to its

dissemination in_ public libraries, without judicial

determination or even the semblance of First Amendment

due process. See generally, Freedman v. Maryland, 380

U.S. 51, 58 (1965) (invalidating film censorship statute as

prior restraint because it lacked several procedural

safeguards necessary “to obviate the dangers of a censorship

system”). Though the three-judge court did not reach

plaintiffs’ prior restraint argument, it warrants analysis as an

* CIPA would clearly fail constitutional scrutiny even under rational

basis review. There may be no better example of irrationality than

mandated government use of a product that secretly categorizes and

blocks a huge amount of speech that comes nowhere close to the type of

content the law was intended to restrict

independent reason for stinking down the statute. The only

other court to consider the constitutionality of mandatory

Internet blocking at a public library invalidated the practice

as a prior restraint. See Mainstream Loudoun, 24 F. Supp.

2d at 570 (because mandatory blocking policy “has neither

adequate standards nor adequate procedural safeguards,” it is

an unconstitutional prior restraint). By mandating the use of

blocking programs that block speech that is not even close to

the line between protected and unprotected speech, CIPA

imposes a classic system of prior restraint which

presumptively violates the Constitution. See Near v.

Minnesota, 283 U.S. 697, 713 (1931) (stating that “the chief

purpose of the [First Amendment] is to prevent previous

restraints upon publication”). Blocking programs function

literally as automated censors, blocking speech in advance of

any judicial determination that it is unprotected. They

arbitrarily and irrationally block thousands of Web pages

that are fully protected. In this Court’s words, “[t]his is . . .

the essence of censorship.” Jd.

A postmaster who opened all letters and refused to

deliver letters with the word “sex” in them would clearly be

violating the First Amendment’s rule against prior restraints.

See Blount v. Rizzi, 400 U.S. 410 (1971) (stnking down

statute that allowed Postmaster General to halt use of mail

for commerce in allegedly obscene materials). As the Court

has explained, “[t]he United States may give up the Post

Office when it sees fit, but while it carries it on the use of the

mails is almost as much a part of free speech as the right to

use our tongues.” /d. at 416 (citations omitted). Similarly,

having chosen to fund Internet access, the government “may

not thereafter selectively restrict certain categories of

Internet speech because it disfavors their content.” See

Mainstream Loudoun, 2 F. Supp. 2d at 795-96.

“(A]ny system of prior restraints of expression comes

to [the court] bearing a heavy presumption against its

constitutional validity.” Bantam Books, 372 U.S. at 70

(morality commission, whose purpose was to recommend

prosecution of obscenity, imposed unconstitutional prior

restraint by sending notices to booksellers that certain books

were objectionable); see also Conrad, 420 U.S. at 559

(municipal board’s denial of permission for performance of

the rock musical “Hair” at a city auditorium, because of

reports that the musical was “obscene,” was an

unconstitutional prior restraint); Drive In Theatres, Inc. v.

Huskey, 435 F.2d 228, 230 (4th Cir. 1970) (invalidating as

“unconstitutional prior administrative restraint” a sheriff's

practice of seizing and terminating exhibition of R-rated

movies).

The government argues that a library’s use of filters

is no more a prior restraint than a library’s decision not to

subscribe to Playboy. But CIPA is a federal mandate for all

libraries nationwide who participate in much-needed funding

programs, not a local library book collection decision.

The government once again argues that the speech

can be obtained elsewhere. Once again, this court has

rejected that argument. Reno, 521 U.S. at 880. Surely

Dallas could not have defended its movie censorship on the

grounds that the movie could be seen in Houston or, at a later

time, it was available on video. Interstate Circuit v. Dallas,

390 U.S. 676 (1968). In addition, for many Americans, their

only means of accessing the Internet is at a library. Section

Al, supra.

In marked contrast to the local book selection

process, CIPA cuts libraries out of the decision-making

process entirely. In fact, by delegating the authority to

restrict speech to third-party, non-governmental actors who

will not reveal what they are censoring, or the rules being

used, CIPA compounds the constitutional infirmities inherent

in any prior restraint. Additionally, the government's

argument ignores the public forum qualities of the Internet

even in public libraries. There is no question that the

decisions of blocking programs “to list particular

publications as objectionable do not follow judicial

determinations that such publications may lawfully be

banned.” Bantam Books, 372 U.S. at 70. CIPA’s disabling

provisions inflict further First Amendment injury by vesting

librarians with unbridled discretion to undo selectively the

blocking companies’ censorship decisions. See Forsyth

County v. Nationalist Movement, 505 U.S. 123, 133 (1992)

(“The First Amendment prohibits the vesting of such

unbridled discretion in a government official”);

Shuttlesworth v. City of Birmingham, 394 U.S. 147, 150-51

(1969) (“a law subjecting the exercise of First Amendment

freedoms to the prior restraint of a license” is

unconstitutional absent “narrow, objective, and definite

standards to guide the licensing authority”). CIPA’s

censorship system comes nowhere close to the judicial

review that is required when First Amendment rights are at

stake. See, e.g., Freedman, 380 U.S. at 58-59.

Iii. CIPA Imposes an Unconstitutional Condition

on Funding of Internet Access at Public

Libraries. .

CIPA is also an unconstitutional condition on the

receipt of a federal benefit. See Board of County Comm'rs v.

Umbehr, 518 U.S. 668, 674 (1996); Perry v. Sindermann,

408 U.S. 593, 597 (1972). The statute requires any library

that participates in the e-rate or LSTA programs to install

blocking software on “any of its computers with Internet

access” during “any use of such computers.” 20 U.S.C.

9134(f)(1)(B); 47 U.S.C. § 254(h)(6)(C) (emph. added).

CIPA requires libraries to block speech even where Internet

terminals are wholly funded by non-federal dollars. This

Court invalidated exactly that practice in League of Women

Voters, holding that the government may not use its spending

power to bar grantees “from using even wholly private funds

to finance its” expressive activity. 468 U.S. at 400.

Although the three-judge court did not rule on

plaintiffs’ unconstitutional conditions claim, it included a

lengthy footnote analyzing the relevant cases. J.S. App.

180a-188a n.36. Because “the First Amendment is not

phrased in terms of who holds the right, but rather what is

protected,” J.S. App. at 183a n.36, plaintiffs clearly have a

valid unconstitutional conditions claim based on the First

Amendment nghts of their patrons.” See, e.g., Procunier v.

Martinez, 416 U.S. 396, 409 (1974) (recognizing that First

Amendment protects the right to receive information);

Virginia v. American Booksellers Ass'n, 484 U.S. 383, 392-

93 (1988) (booksellers have standing to sue on behalf of

their customers); J.S. App. 188a n.36. As this Court has

explained, “the question must be whether [the government

is] abridg[ing] expression that the First Amendment was

meant to protect.” First Nat'l Bank of Boston v. Bellotti, 435

U.S. 765, 776 (1978). Contrary to the government’s

argument, then, it is irrelevant that CIPA imposes its

conditions on libraries rather than on their patrons directly.

Govt. Br. at 41. This Court has previously entertained

unconstitutional conditions claims both by the organizations

that receive federal funding and their constituents. See e.g.,

Velazquez, 531 U.S. at 537 (lawyers and clients sued to

declare restriction on advocacy invalid); Rust v. Sullivan,

500 U.S. 173, 181 (1991) (doctors and patients sued to

invalidate restriction on use of Title X funds); League of

Women Voters, 468 U.S. at 370 n.6 (station owner and

viewers sued to invalidate conditions on station’s receipt of

federal funds).

* Thus, this Court need not decide whether public libraries, as municipal

entities, have their own independent First Amendment nghts.

In Velazquez, the Court invalidated a restriction on

funding because the program funded private speech, not

governmental speech, and because the restriction distorted

the traditional function of the medium. 531 U.S. at 544-45.

Similarly, CIPA distorts the usual function of public libraries

to provide uncensored access to information. The Legal

Services Corporation (“LSC”) was established to distribute

federal funds to eligible local grantee organizations “for the

purpose of providing financial support for legal assistance in

noncriminal proceedings or matters to persons financially

unable to afford legal assistance.” Jd. at 536. Congress,

however, prohibited legal representation funded by recipients

of LSC moneys if the representation involved an effort to

amend or otherwise challenge existing welfare law. Jd.

The Court held that this condition on the use of LSC

funds violated the First Amendment rights of LSC grantees

and their clients.° When the government “designs a program

to facilitate private speech, and not to promote a govern-

mental message,” and “seeks to use an existing medium of

expression and to control it . . . in ways which distort its

usual functioning,” the distorting restriction must be struck

down. /d. at 534 (“[rJjestricting LSC attorneys in advising

their clients and in presenting arguments and analyses to the

courts distorts the legal system by altering the traditional role

of the attorneys”). See also League of Women Voters, 468

U.S. at 396-97 (prohibitions against editorializing by public

° Velasquez also condemned the program because it required viewpoint

discrimination. 531 U.S. at 533, citing Board of Regents v. Southworth,

529 U.S. 217, 229 (2000). Because blocking programs make their

censorship decisions in secret, there is no way to know whether they are

viewpoint discriminatory. For example, blocking programs may block

content targeted to the gay and lesbian community not because it is

actually sexually explicit, but because the vendors disapprove of that

lifestyle. The potential for viewpoint discrimination certainly adds to its

constitutional defects. See Mainstream Loudoun, 24 F. Supp. 2d at 565-

67.

radio networks were an impermissible restriction, even

though the government enacted the restriction to control the

use of public funds; First Amendment forbade the

government from using the forum in an unconventional way

to suppress speech inherent in the nature of the medium).

When the government conversely “establishes a subsidy for

specified ends,” “certain restrictions may be necessary to

define the limits and purposes of the program.” /d. (citing,

inter alia, Perry Ed. Assn. v. Perry Local Educators’ Assn.,

460 U.S. 37, (1983); Rust v. Sullivan, 500 U.S. 173 (1991)).

CIPA seeks to distort an existing medium of

expression at public libraries by requiring them to:

(1) deny patrons access to constitutionally

protected speech that libraries would

otherwise provide to patrons; and (2) delegate

decision making to private’ software

developers who closely guard their selection

criteria as trade secrets and who do not

purport to make their decisions on the basis of

whether the blocked Web sites are

constitutionally protected or would add value

to a public library’s collection.

J.S. App. 187a n.36. Like the improper effect of the funding

restriction this Court invalidated in Velazquez, “(bly

interfering with public libraries’ discretion to make available

to patrons as wide a range of constitutionally protected

speech as possible, the federal government is arguably

distorting the usual functioning of public libraries as places

of freewheeling inquiry.” J.S. App. 187a n.36.

CONCLUSION

For the reasons stated above, the judgment of the

three judge court should be affirmed.

Respectfully Submited,

Christopher A. Hansen

(Counsel of Record)

Ann Beeson

Steven R. Shapiro

American Civil Liberties Union

Foundation

125 Broad Street

New York, New York 10004

(212) 549-2500

Stefan Presser

ACLU of Pennsylvania

125 South Ninth Street, Suite 701

Philadelphia, Pennsylvania 19107

(215) 592-1513

Charles S. Sims

Proskauer Rose LLP

1585 Broadway

New York, New York 10036-8299

(212) 969-3000

David L. Sobel

Electronic Privacy Information Center

1718 Connecticut Avenue NW,

Suite 200

Washington, D.C. 20009

(202) 483-1140

Lee Tien

Electronic Frontier Foundation

454 Shotwell Street

San Francisco, California 94110

(415) 436-9333

FOR PLAINTIFF MULTNOMAH

COUNTY PUBLIC LIBRARY

Thomas Sponsler, County Attorney

Multnomah County

501 N.E. Hawthorne Blvd.,

Suite 500

Portland, OR 97214

(503) 988-3138

Dated: February 10, 2003

This is a copy of a public record, reproduced as it was published. It is not legal advice, and it may not be the version a court would rely on. Check the official source before you cite it.