Barrett v. Rosenthal

Opinion

In the Communications Decency Act of 1996, Congress declared: “No provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider.” (47 U.S.C. § 230(c)(1).)1 “No cause of action may be brought and no liability may be imposed under any State or local law that is inconsistent with this section.” (§ 230(e)(3).)

These provisions have been widely and consistently interpreted to confer broad immunity against defamation liability for those who use the Internet to publish information that originated from another source. The immunity has been applied regardless of the traditional distinction between “publishers” and “distributors.” Under the common law, “distributors” like newspaper vendors and booksellers are hable only if they had notice of a defamatory statement in their merchandise. The publisher of the newspaper or book where the statement originally appeared, however, may be held liable even without notice.

In this case, the Court of Appeal diverged from the prevailing interpretation of section 230. It decided that common law “distributor” liability survived the congressional grant of immunity, so that Internet service providers and users are exposed to liability if they republish a statement with notice of its defamatory character.

We granted review to decide whether section 230 confers immunity on “distributors.” Because this case involves the liability of an individual rather than a service provider, we asked the parties to address the definition of the statutory term “user.” We also requested briefing on whether the immunity analysis is affected if a user engages in active rather than passive conduct. We *40conclude that section 230 prohibits “distributor” liability for Internet publications. We further hold that section 230(c)(1) immunizes individual “users” of interactive computer services, and that no practical or principled distinction can be drawn between active and passive use. Accordingly, we reverse the Court of Appeal’s judgment.

We acknowledge that recognizing broad immunity for defamatory republications on the Internet has some troubling consequences. Until Congress chooses to revise the settled law in this area, however, plaintiffs who contend they were defamed in an Internet posting may only seek recovery from the original source of the statement.

I. FACTUAL AND PROCEDURAL BACKGROUND

Plaintiffs, Dr. Stephen J. Barrett and Dr. Terry Polevoy, operated Web sites devoted to exposing health frauds. Defendant llena Rosenthal directed the Humantics Foundation for Women and operated an Internet discussion group. Plaintiffs alleged that Rosenthal and others committed libel by maliciously distributing defamatory statements in e-mails and Internet postings, impugning plaintiffs’ character and competence and disparaging their efforts to combat fraud.2 They alleged that Rosenthal republished various messages even after Dr. Barrett warned her they contained false and defamatory information.

Rosenthal moved to strike the complaint under the anti-SLAPP statute. (Code Civ. Proc., § 425.16; SLAPP is an acronym for strategic lawsuit against public participation.) She claimed her statements were protected speech, and argued that plaintiffs could not establish a probability of prevailing because she was immune under section 230. (See Code Civ. Proc., § 425.16, subd. (b); Equilon Enterprises v. Consumer Cause, Inc. (2002) 29 Cal.4th 53, 67 [124 Cal.Rptr.2d 507, 52 P.3d 685].) She also contended her statements were not actionable.

The court granted the motion, finding that Rosenthal’s statements concerned an issue of public interest within the scope of the anti-SLAPP statute, *41and were, for the most part, not actionable because they contained no provably false assertions of fact. Plaintiffs do not challenge that ruling. The court determined that the only actionable statement appeared in an article Rosenthal received via e-mail from her codefendant Tim Bolen. This article, subtitled “Opinion by Tim Bolen,” accused Dr. Polevoy of stalking a Canadian radio producer. Rosenthal posted a copy of this article on the Web sites of two newsgroups devoted to alternative health issues and the politics of medicine, not on the site of her own discussion group. According to Rosenthal, these newsgroups were part of “the wild west of the Internet,” with “no administrators and no one to enforce rules of conduct.”3 The trial court ruled that this republication was immunized by section 230(c)(1).

The Court of Appeal vacated the order granting the motion to strike insofar as it applied to Dr. Polevoy. It held that section 230 did not protect Rosenthal from liability as a “distributor” under the common law of defamation. We granted Rosenthal’s petition for review.4

II. DISCUSSION

The leading case on section 230 immunity rejected the “distributor” liability theory adopted by the Court of Appeal here. (Zeran v. America Online, Inc. (4th Cir. 1997) 129 F.3d 327, 331-333 (Zeran).) We first discuss the Zeran holding and rationale, then the Court of Appeal’s contrary analysis.5 Recognizing “distributor” liability would have a dramatic impact on *42Internet service providers. We agree with the Zeran court that Congress did not intend to create such an exception to section 230 immunity.6

*43Rosenthal, however, is not a service provider, at least with respect to the newsgroups where she posted the Bolen article. This appears to be the first published case in which section 230 immunity has been invoked by an individual who had no supervisory role in the operation of the Internet site where allegedly defamatory material appeared, and who thus was clearly not a provider of an “interactive computer service” under the broad definition provided in the CDA. (§ 230(f)(2); see fn. 7, ante.) Accordingly, we asked the parties to brief the meaning of the term “user” in section 230, and whether any distinction might be drawn between active and passive use under the statute. In part C of our discussion, we conclude that Congress employed the term “user” to refer simply to anyone using an interactive computer service, without distinguishing between active and passive use.

A. Zeran

Kenneth Zeran was bombarded with angry and derogatory telephone calls, including death threats, after an unidentified person posted a message on an America Online, Inc. (AOL) bulletin board. The message advertised T-shirts with offensive slogans referring to the Oklahoma City bombing of the Alfred R Murrah Federal Building, and instructed prospective purchasers to call Zeran’s home telephone number. Zeran notified AOL of the problem, and the posting was eventually removed. However, similar postings appeared, and an Oklahoma radio announcer aired the contents of the first message. Zeran was again inundated with threatening phone calls. He sued AOL for unreasonable delay in removing the defamatory messages, refusing to post retractions, and failing to screen for similar postings. (Zeran, supra, 129 F.3d at pp. 328-329.)

AOL successfully moved for judgment on the pleadings, relying on section 230. (Zeran, supra, 129 F.3d at pp. 329-330.) The Fourth Circuit Court of Appeals affirmed, holding that the plain language of section 230 “creates a federal immunity to any cause of action that would make service providers liable for information originating with a third-party user of the service. Specifically, § 230 precludes courts from entertaining claims that would place a computer service provider in a publisher’s role. Thus, lawsuits seeking to hold a service provider liable for its exercise of a publisher’s traditional editorial functions—such as deciding whether to publish, withdraw, postpone or alter content—are barred.” (Zeran, at p. 330.)

*44Referring to the congressional finding that the Internet has flourished “with a minimum of government regulation” (§ 230(a)(4)), and the policy statement favoring a free market for interactive computer services “unfettered by Federal or State regulation” (§ 230(b)(2)), the Zeran court reasoned that Congress viewed “[t]he imposition of tort liability on service providers for the communications of others” as “simply another form of intrusive government regulation of speech.” (Zeran, supra, 129 F.3d at p. 330.) While original posters of defamatory speech do not escape accountability, Congress “made a policy choice . . . not to deter harmful online speech [by] imposing tort liability on companies that serve as intermediaries for other parties’ potentially injurious messages.” (Id. at pp. 330-331.) This policy reflects a concern that if service providers faced tort liability for republished messages on the Internet, they “might choose to severely restrict the number and type of messages posted.” (Id. at p. 331.)

The court noted that another important purpose of section 230 was “to encourage service providers to self-regulate the dissemination of offensive material over their services.” (Zeran, supra, 129 F.3d at p. 331.) The legislative history indicates that section 230 was enacted in response to an unreported New York trial court case. (Stratton Oakmont, Inc. v. Prodigy Services Co. (Sup.Ct. May 24, 1995) 23 Media L.Rep. 1794 [1995 WL 323710] (Stratton Oakmont).)7 There, a service provider was held liable for defamatory comments posted on one of its bulletin boards, based on a finding that the provider had adopted the role of “publisher” by actively screening and editing postings. “Fearing that the specter of liability would . . . deter service providers from blocking and screening offensive material, Congress enacted § 230’s broad immunity,” which “forbids the imposition of publisher liability on a service provider for the exercise of its editorial and self-regulatory functions.” (Zeran, supra, 129 F.3d at p. 331.)

Zeran made the same argument adopted by the Court of Appeal here: that Congress intended to distinguish between “publishers” and “distributors,” immunizing publishers but leaving distributors exposed to liability. At common law, “primary publishers,” such as book, newspaper, or magazine publishers, are liable for defamation on the same basis as authors. Booksellers, news vendors, or other “distributors,” however, may only be held liable if they knew or had reason to know of a publication’s defamatory *45content. (Zeran, supra, 129 F.3d at p. 331; Prosser & Keeton, The Law of Torts (5th ed. 1984) § 113, pp. 810-811; Rest.2d Torts, § 581, subd. (1), & corns, c, d, & e, pp. 232-234; see also Osmond v. EWAP, Inc. (1984) 153 Cal.App.3d 842, 852-854 [200 Cal.Rptr. 674].)8 Zeran contended that because Congress mentioned only the term “publisher” in section 230, it intended to leave “distributors” unprotected. He claimed that once he gave AOL notice that it was posting defamatory statements on its bulletin board, AOL became liable as a “distributor.” (Zeran, supra, 129 F.3d at pp. 331-332.)

The Zeran court held that the publisher/distributor distinction makes no difference for purposes of section 230 immunity. Publication is a necessary element of all defamation claims, and includes every repetition and distribution of a defamatory statement. (Zeran, supra, 129 F.3d at p. 332, citing Prosser & Keeton, The Law of Torts, supra, § 113, pp. 799, 802, 803, and Rest.2d Torts, §§ 558, subd. (b) & 577.) Although “distributors” become liable only upon notice, they are nevertheless included in “the larger publisher category.” (Zeran, supra, 129 F.3d at p. 332.) “Zeran simply attaches too much importance to the presence of the distinct notice element in distributor liability. . . . [Ojnce a computer service provider receives notice of a potentially defamatory posting, it is thrust into the role of a traditional publisher. The computer service provider must decide whether to publish, edit, or withdraw the posting. In this respect, Zeran seeks to impose liability on AOL for assuming the role for which § 230 specifically proscribes liability—the publisher role.” (Id. at pp. 332-333.)

Subjecting service providers to notice liability would defeat “the dual purposes” of section 230, by encouraging providers to restrict speech and abstain from self-regulation. (Zeran, supra, 129 F.3d at p. 333.) A provider would be at risk for liability each time it received notice of a potentially defamatory statement in any Internet message, requiring an investigation of the circumstances, a legal judgment about the defamatory character of the information, and an editorial decision on whether to continue the publication. “Although this might be feasible for the traditional print publisher, the sheer number of postings on interactive computer services would create an impossible burden in the Internet context.” (Ibid.)

“Similarly, notice-based liability would deter service providers from regulating the dissemination of offensive material over their own services. Any efforts by a service provider to investigate and screen material posted on its *46service would only lead to notice of potentially defamatory material more frequently and thereby create a stronger basis for liability. Instead of subjecting themselves to further possible lawsuits, service providers would likely eschew any attempts at self-regulation.

“More generally, notice-based liability for interactive computer service providers would provide third parties with a no-cost means to create the basis for future lawsuits. Whenever one was displeased with the speech of another party conducted over an interactive computer service, the offended party could simply ‘notify’ the relevant service provider, claiming the information to be legally defamatory. . . . Because the probable effects of distributor liability on the vigor of Internet speech and on service provider self-regulation are directly contrary to § 230’s statutory purposes, we will not assume that Congress intended to leave liability upon notice intact.” (Zeran, supra, 129 F.3d at p. 333.)

In support of his argument for notice-based liability, Zeran invoked the rule against abrogation of common law principles unless Congress speaks directly to the question. (Zeran, supra, 129 F.3d at pp. 333-334; United States v. Texas (1993) 507 U.S. 529, 534 [123 L.Ed.2d 245, 113 S.Ct. 1631].) However, the court reasoned that Congress had spoken directly by employing the term “publisher,” and that preserving “distributor” liability would defeat the primary purposes of section 230. The policy of strictly construing statutes in derogation of the common law does not require a literal interpretation conflicting with the obvious legislative purpose. (Zeran, supra, 129 F.3d at p. 334, citing Isbrandtsen Co. v. Johnson (1952) 343 U.S. 779, 783 [96 L.Ed. 1294, 72 S.Ct. 1011].)

The Zeran court’s views have been broadly accepted, in both federal and state courts.9 Before the Court of Appeal issued its opinion below, two other California Courts of Appeal had followed Zeran. In Kathleen R. v. City of Livermore (2001) 87 Cal.App.4th 684 [104 Cal.Rptr.2d 772], a taxpayer sued after her son obtained sexually explicit photographs through an Internet *47connection at a public library. She sought injunctive relief on various theories of liability. (Id. at pp. 690-691.) The Kathleen R. court held that the state law causes of action were barred by section 230. (Kathleen R., at p. 692.) It cited Zeran for the rule that section 230(c)(1) immunizes both “publisher[s]” and “distributor[s].” (Kathleen R., at p. 695, fn. 3.) It also agreed with the Zeran court’s analysis of congressional intent. (Id. at p. 697.)

In Gentry v. eBay, Inc. (2002) 99 Cal.App.4th 816 [121 Cal.Rptr.2d 703], the plaintiffs used eBay’s on-line marketing services to purchase sports memorabilia. Claiming the items bore forged autographs, they sued eBay for negligence, unfair trade practices, and violation of Civil Code section 1739.7, which regulates the sale of such collectibles. (Gentry, at p. 820.) The Gentry court ruled that section 230 immunized eBay from liability on all the plaintiffs’ claims. It noted the broad scope given to section 230 immunity by the Zeran court and others, and reasoned that the plaintiffs were trying to hold eBay responsible for disseminating information provided by the individual sellers who used its service. (Gentry, at pp. 828-831.) Regarding the allegation that eBay knew or should have known about the sellers’ illegal conduct but failed to prevent it by withdrawing or altering the fraudulent content, the Gentry court stated; “This is the classic kind of claim that Zeran found to be preempted by section 230, . . . one that seeks to hold eBay liable for its exercise of a publisher’s traditional editorial functions.” (Id. at p. 835.)

B. The Court of Appeal Analysis

Swimming against the jurisprudential tide, the Court of Appeal in this case disputed the ruling in Zeran, contending it confers a more expansive immunity than is necessary to preserve freedom of online speech, and would actually defeat the goal of encouraging self-regulation. The Court of Appeal focused on three factors: (1) the Zeran court’s interpretation of the statutory term “publisher;” (2) the legislative history of section 230; and (3) the practical implications of notice liability in the Internet environment. We reject the Court of Appeal’s analysis on each of these points.

1. The Meaning of “Publisher”

The Court of Appeal acknowledged that publication is an element of defamation, and that “distributors” are sometimes referred to as “secondary publishers.” (See, e.g., Dworkin v. Hustler Magazine, Inc. (D.C.Wyo. 1985) 611 F.Supp. 781, 785; Hart v. Bennet (2003) 2003 WI App 231 [267 Wis.2d 919, 672 N.W.2d 306, 318, fn. 14]; Prosser & Keeton, The Law of Torts, supra, § 113, at p. 803; Smolla, The Law of Defamation (2d ed. 2005) § 4:92, p. 4-140.15.) However, the court pronounced it “reasonable to assume” that Congress had in mind the different standards of common law liability *48imposed on “primary publishers,” who have control over content, and “distributors,” who do not. Thus, the omission of any reference to “distributors” in section 230(c)(1) was arguably intentional.

The Court of Appeal noted that the goal of discouraging excessive self-censorship by immunizing publishers is at odds with the rights of individuals to recover for defamatory falsehood. It deemed the term “publisher” ambiguous, because it might refer to primary publishers only or to both primary publishers and distributors. According to the Court of Appeal, such a “legally uncertain word” could not support the broad immunity the Zeran court derived from the statute. It found nothing in the statutory findings and declarations to indicate that Congress considered online speech in need of blanket protection. Indeed, it detected a contrary intent in the terms of section 230(c)(2), which immunizes providers and users against liability for “any action voluntarily taken in good faith to restrict access to or availability of material that the provider or user considers to be . . . objectionable, whether or not such material is constitutionally protected” or to provide others with “the technical means to restrict access to [such] material.” The Court of Appeal reasoned that section 230(c)(2) would be superfluous if all “publishers” enjoyed absolute immunity under section 230(c)(1).

The Court of Appeal sought further support for limiting the scope of the term “publisher” to primary publishers by comparing the immunity provisions of the CDA with those of the Digital Millenium Copyright Act, enacted in 1998 (DMCA; 17 U.S.C. § 512).10 The DMCA immunizes Internet service providers from liability for copyright infringement if the provider is unaware of the infringement and acts expeditiously to remove the copyrighted material upon notice. It includes detailed notice requirements, and procedures for replacement of the disputed material upon sufficient counter-notification. (17 U.S.C. § 512(c) & (g).) Because Congress did not include such specific regulation of notice liability in the CDA,. the Court of Appeal decided it had failed to “speak directly” to the issue, thus preserving common law distributor liability. (See United States v. Texas, supra, 507 U.S. at p. 534.)

We conclude the Zeran court’s construction of the term “publisher” is sound. The terms of section 230(c)(1) are broad and direct: “No provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider.” Given that “distributors” are also known as “secondary publishers,” there is little reason to believe Congress felt it necessary to address them separately. There is even less reason to suppose that Congress intended to immunize “publishers” but leave “distributors” open to liability, when the responsibility of publishers for offensive content is greater than that of mere *49distributors. The Court of Appeal failed to respond to the Zeran court’s point that once online distributors are notified of defamatory content, they are placed in a position traditionally occupied by publishers, and must make an editorial decision on how to treat the posted material. (Zeran, supra, 129 F.3d at p. 332.) This is a persuasive justification for giving the term “publisher” an inclusive interpretation. (See, e.g., Gentry v. eBay, Inc., supra, 99 Cal.App.4th at p. 835; Green v. America Online, supra, 318 F.3d at p. 471; Donato v. Moldow, supra, 865 A.2d at pp. 725-726; Schneider v. Amazon.com, Inc., supra, 31 P.3d at pp. 41-42.)

We are not convinced by the Court of Appeal’s reasoning that a broad reading of section 230(c)(1) would make section 230(c)(2) unnecessary. These provisions address different concerns. Section 230(c)(1) is concerned with liability arising from information provided online. Section 230(c)(2) is directed at actions taken by Internet service providers or users to restrict access to online information.11 Liability for censoring content is not ordinarily associated with the defendant’s status as “publisher” or “speaker.” Those terms, employed in section 230(c)(1), are drawn from the law of defamation. (See, e.g., Prosser & Keeton, The Law of Torts, supra, § 113, at p. 803; Rest.2d Torts, § 568.) Section 230(c)(1) provides immunity from claims by those offended by an online publication, while section 230(c)(2) protects against claims by those who might object to the restriction of access to an online publication.

The Court of Appeal’s reference to the DMCA does not support its conclusion that Congress’s use of the term “publisher” was insufficient to abrogate “distributor” liability. To the contrary, the DMCA shows that Congress has crafted a limited immunity in a closely related context, with specific provision for notice liability. (17 U.S.C. § 512(c).) The fact that it did not do so in the CDA, and has not amended section 230 to add a similar provision in the 10 years since it was enacted, or in the eight years since the example of the DMCA has been in existence, strongly supports the conclusion that Congress did not intend to permit notice liability under the CDA.12

We note that it is far from clear how the distinction between traditional print publishers and distributors would apply in the Internet environment, *50with its many and various forms of discourse. (See Reno v. American Civil Liberties Union, supra, 521 U.S. 844, 850-853.) As the high court noted, “[a]ny person or organization with a computer connected to the Internet can ‘publish’ information.” (Id. at p. 853.) Whenever such information is copied from another source, its publication might also be described as a “distribution.” The distinction proposed by the Court of Appeal, based on rules developed in the post-Gutenberg, pre-cyberspace world, would foster disputes over which category the defendant should occupy. The common law of defamation would provide little guidance.

In this case, for example, Rosenthal could claim that her active role in selecting and posting material disparaging plaintiffs qualified her as a primary publisher. Her participation in the dissemination of the Bolen article, particularly considered in light of her other alleged verbal attacks on plaintiffs, arguably went beyond mere distribution. (See Prosser & Keeton, The Law of Torts, supra, § 113, at p. 803; Smolla, The Law of Defamation, supra, § 4:92, p. 4-140.15; Rest.2d Torts, § 568.) The Court of Appeal provided no analysis justifying its conclusion that Rosenthal could be held liable as a “distributor,” noting only that she alleged no facts preventing her from being so characterized. We need not decide the question, but certainly the argument could be made that plaintiffs’ allegations cast Rosenthal in the role of a “publisher.”

2. The Legislative History

The Court of Appeal noted that section 230 was enacted along with other CDA provisions that prohibited the knowing transmission of “obscene or indecent” or “patently offensive” messages to persons under the age of 18.13 It reasoned that immunizing Internet service providers and users from “primary publisher” liability advanced a similar purpose by protecting those providers and users who try but fail to identify and remove offensive material. However, according to the Court of Appeal, immunization from “distributor” liability would be inconsistent with this goal because it would protect providers and users who make no effort to screen for offensive material, along with those who refuse to take action once on notice.

The Court of Appeal claimed support for this view in the legislative history of section 230, though it conceded that the history is “meager.” (See Sheridan, Zeran v. AOL and the Effect of Section 230 of the Communications Decency Act Upon Liability for Defamation on the Internet (1997) 61 *51Alb. L.Rev. 147, 168 (hereafter Sheridan).) The court recognized that section 230 was enacted to remove the disincentives to self-regulation created by the Stratton Oakmont case, in which a service provider was held liable as a primary publisher because it actively screened and edited messages posted on its bulletin boards. (Stratton Oakmont, supra, 23 Media L.Rep. 1794 [1995 WL 323710]; see Zeran, supra, 129 F.3d at p. 331; fn. 7, ante.) However, the Court of Appeal considered an earlier Internet defamation case to be equally important in ascertaining the purpose of section 230.

In Cubby, Inc. v. CompuServe, Inc. (S.D.N.Y. 1991) 776 F.Supp. 135, a journalist claimed he was defamed by a competitor’s remarks posted on an Internet forum provided by CompuServe. (Id. at pp. 137-138.) The court applied the common law “distributor” standard of liability, concluding the forum was essentially an electronic library over which CompuServe exercised little or no editorial control. (Id. at pp. 139-140.) Because there was no evidence CompuServe knew or had reason to know of the statements, the court granted it summary judgment. (Id. at p. 141.)

The Court of Appeal noted that Cubby was distinguished in Stratton Oakmont, and also in comments by the sponsors of section 230. As related in a law review article relied on by the court, “Representative Cox, one of two sponsors of the immunity provision, characterized the imposition of distributor liability in Cubby as holding that CompuServe ‘was not the publisher or editor’ of the material. He clearly used the term ‘publisher’ to exclude parties held to the distributor liability standard applied to CompuServe in that case. 141 Cong. Rec. H8469 (daily ed. Aug. 4, 1995) (statement of Rep. Cox). The provision’s sponsors summarized both the Cubby and Stratton [Oakmont] decisions, and then repeatedly discussed the need to overrule Stratton [Oakmont], without again mentioning Cubby. See [141 Cong. Rec. H8469 (daily ed. Aug. 4, 1995) (statements of Rep. Cox and Rep. Wyden]; see also 141 Cong. Rec. S8345 (daily ed. June 14, 1995) (statements of Sen. Coats) (distinguishing between publisher and distributor liability and noting that the [CDA] was designed not to hold intermediaries to publisher liability).” (Freiwald, Comparative Institutional Analysis in Cyberspace: The Case of Intermediary Liability for Defamation (2001) 14 Harv. J.L. & Tech. 569, 632, fn. 259 (hereafter Freiwald).)

From these sources, the Court of Appeal discerned a congressional intent to preserve “distributor” liability. It cited several academic commentators for the view that immunizing Internet service providers from “distributor” liability would actually frustrate the objective of self-regulation, because no liability would flow from failing to screen for defamatory content. (McManus, Rethinking Defamation Liability for Internet Service Providers (2001) 35 Suffolk U. L.Rev. 647, 668 (hereafter McManus); Patel, Immunizing Internet *52Service Providers From Third Party Internet Defamation Claims: How Far Should Courts Go? (2002) 55 Vand. L.Rev. 647, 684; see also Sheridan, supra, 61 Alb. L.Rev. at pp. 169-170.)

The Court of Appeal and the commentators on which it relied read too much into the legislative record. We note that the comments of Senator Coats, summarized by Professor Freiwald as quoted above, pertained not to section 230 but to a separate provision of the CD A, codified at title 47 United States Code section 223(f)(4). (141 Cong.Rec. S8328, S8345 (daily ed. June 14, 1995).) The comments of Representative Cox, a sponsor of section 230, are pertinent but do not indicate that distributors were meant to be excluded from statutory protection.14

Representative Cox said section 230 was intended to “encourage people like . . . CompuServe . . . [by] ... [][]... [H protecting] them from taking *53on liability such as occurred in the [Stratton Oakmont] case in New York that they should not face for helping us . . . solve this problem.” (141 Cong. Rec. H8470 (daily ed. Aug. 4, 1995).) Thus, he meant that “distributors” like CompuServe would be protected from rather than threatened with liability, to encourage responsible screening of the content provided on their services. Under the Court of Appeal’s interpretation of section 230, a “distributor” could be sued if it deleted material after receiving notice of offensive content, but did not act quickly or thoroughly enough to suit the offended party. Primary “publishers” who decide not to remove offensive postings would be immunized, while “distributors” making the same decision would be unprotected. It is unlikely that Congress intended such incongruous results.

Both the terms of section 230(c)(1) and the comments of Representative Cox reflect the intent to promote active screening by service providers of online content provided by others. Congress implemented its intent not by maintaining the common law distinction between “publishers” and “distributors,” but by broadly shielding all providers from liability for “publishing” information received from third parties.15 Congress contemplated self-regulation, rather than regulation compelled at the sword point of tort liability. It chose to protect even the most active Internet publishers, those who take an aggressive role in republishing third party content. It would be anomalous to hold less active “distributors” liable upon notice. Thus, the immunity conferred by section 230 applies even when self-regulation is unsuccessful, or completely unattempted. (Blumenthal v. Drudge, supra, 992 F.Supp. at p. 52; Schneider v. Amazon.com, Inc., supra, 31 P.3d at p. 43; Donato v. Moldow, supra, 865 A.2d at p. 726.)

*54As Rosenthal and amici curiae point out, subsequent legislative history contains explicit support for the Zeran court’s interpretation. In 2002, Congress enacted the Dot Kids Implementation and Efficiency Act.16 A House committee report notes that the purpose of this legislation was “to facilitate the creation of a new, second-level Internet domain within the United States country code domain that will be a haven for material that promotes positive experiences for children and families using the Internet.” (H.R. Rep. No. 107-449, 2d. Sess., p. 5 (2002).) The legislation includes a provision that the new registry it created, and related entities, “are deemed to be interactive computer services for purposes of section 230(c) of the Communications Act of 1934 (47 U.S.C. 230(c)).” (47 U.S.C. § 941(e)(1).) The committee report explains that this provision was “intended to shield the Mrids.us’ registry, registrars, and parties who contract with the registry, from liability based on self-policing efforts to intercept and take down material that is not ‘suitable for minors’ or is ‘harmful to minors.’ The Committee notes that ISPs [Internet service providers] have successfully defended many lawsuits using section 230(c). The courts have correctly interpreted section 230(c), which was aimed at protecting against liability for such claims as negligence[.] (See, e.g., Doe v. America Online, 783 So.2d 1010 (Fla. 2001)) and defamation (Ben Ezra, Weinstein, and Co. v. America Online, 206 F.3d 980 (2000); Zeran v. America Online, 129 F.3d 327 (1997)). The Committee intends these interpretations of section 230(c) to be equally applicable to those entities covered by H.R. 3833.”17 (H.R. Rep. No. 107-449, supra, p. 13.)

3. Practical Implications of Notice Liability

The Zeran court identified three deleterious effects that would flow from reading section 230 to permit liability upon notice. First, service providers who received notification of a defamatory message would be subject to liability only for maintaining the message, not for removing it. This fact, together with the burdens involved in evaluating the defamatory character of a great number of protested messages, would provide a natural *55incentive to simply remove messages upon notification, chilling the freedom of Internet speech. Second, notice-based liability would deter service providers from actively screening the content of material posted on its service, because discovering potentially defamatory material would only increase the provider’s liability. Finally, notice-based liability would give third parties a cost-free means of manufacturing claims, imposing on providers “ceaseless choices of suppressing controversial speech or sustaining prohibitive liability.” (Zeran, supra, 129 F.3d at p. 333.)

The Court of Appeal expressed doubt that a statute encouraging service providers to restrict access to offensive material was intended to promote free speech over the Internet. It also questioned the “speculative conclusion” that notice-based liability would significantly chill online speech, though it refrained from taking a definitive position on this point. Noting the absence of any evidence in the record regarding the burdens such liability would create, the Court of Appeal referred to the views of commentators critical of Zeran as a way to explore the contours of the debate without attempting its resolution. (See Intel Corp. v. Hamidi (2003) 30 Cal.4th 1342, 1363 [1 Cal.Rptr.3d 32, 71 P.3d 296].)

Some critics have suggested that market forces would restrain service providers from removing postings without investigation, because any provider engaging in that practice would acquire a bad reputation in the Internet community. (Sheridan, supra, 61 Alb. L.Rev. at p. 176; Freiwald, supra, 14 Harv. J.L. & Tech, at p. 622; Butler, Plotting the Return of an Ancient Tort to Cyberspace: Towards a New Federal Standard of Responsibility for Defamation for Internet Service Providers (1999-2000) 6 Mich. Telecomm. & Tech. L.Rev. 247, 264.) It has also been argued that the difficulty of prevailing on a defamation claim would attenuate the burden of notice-based liability on providers. Moreover, because “distributor” liability would only arise upon notice, and would not require service providers to review postings in advance, defamation damages would be limited to those accruing after the provider became aware of the defamatory character of a message. (Sheridan, supra, 61 Alb. L.Rev. at p. 173.)

Citing McManus, supra, 35 Suffolk U. L.Rev. at page 661, the Court of Appeal asserted that Zeran has been criticized for failing to account for the many different ways defamation may be transmitted over the Internet, and the different levels of control an Internet intermediary may exercise over the content of messages. Most fundamentally, however, the Court of Appeal noted that critics have condemned Zeran for giving insufficient consideration to the interests of defamation victims. American courts have striven to develop rules that balance the legitimate protections of defamation liability with the constitutional right to free speech. The Court of Appeal resisted the *56notion that a blanket immunity derived from section 230(c)(1) should disturb that balance. It concluded that preserving “distributor” liability was consistent with the immunity provisions of section 230.

The Court of Appeal gave insufficient consideration to the burden its rule would impose on Internet speech. It is inaccurate to suggest that Congress was indifferent to free speech protection when it enacted section 230. The statute includes findings welcoming the “extraordinary advance in the availability of educational and informational resources” on the Internet, and applauding the Internet as a “forum for a true diversity of political discourse” that offers “myriad avenues for intellectual activity” and provides “a variety of political, educational, cultural, and entertainment services.” (§ 230(a)(1), (3), & (5).) Congress sought to “promote the continued development of the Internet and other interactive computer services.” (§ 230(b)(1).) The provisions of section 230(c)(1), conferring broad immunity on Internet intermediaries, are themselves a strong demonstration of legislative commitment to the value of maintaining a free market for online expression.

The fact that Congress also meant to restrict access to certain Internet content does not compel a contrary conclusion. As the court aptly observed in Batzel v. Smith, supra, 333 F.3d 1018; “[T]here is an apparent tension between Congress’s goals of promoting free speech while at the same time giving parents the tools to limit the material their children can access over the Internet. As a result of this apparent tension, some commentators have suggested that the Fourth Circuit in Zeran imposed . . . First Amendment goals on legislation that was actually adopted for the speech-restrictive purpose of controlling the dissemination of content over the Internet. [Citation.] These critics fail to recognize that laws often have more than one goal in mind, and that it is not uncommon for these purposes to look in opposite directions. The need to balance competing values is a primary impetus for enacting legislation. Tension within statutes is often not a defect but an indication that the legislature was doing its job.” (Id. at p. 1028; see also Carafano v. Metrosplash.com, Inc., supra, 339 F.3d at pp. 1122-1123.)

We agree with the Zeran court, and others considering the question, that subjecting Internet service providers and users to defamation liability would tend to chill online speech. (See Carafano v. Metrosplash.com, Inc., supra, 339 F.3d at pp. 1123-1124; Batzel v. Smith, supra, 333 F.3d at pp. 1027-1028; Noah v. AOL Time Warner, Inc. (ED Va. 2003) 261 F.Supp.2d 532, 538; Blumenthal v. Drudge, supra, 992 F.Supp. at p. 52; Donato v. Moldow, supra, 865 A.2d at p. 726.) Certainly, that conclusion is no more speculative than the surmise that market forces might deter providers from removing postings without investigating their defamatory character.

*57We reject the argument that the difficulty of prevailing on a defamation claim mitigates the deterrent effect of potential liability. Defamation law is complex, requiring consideration of multiple factors. These include whether the statement at issue is true or false, factual or figurative, privileged or unprivileged, whether the matter is of public or private concern, and whether the plaintiff is a public or private figure. (See 5 Witkin, Summary of Cal. Law (10th ed. 2005) Torts, §§ 529, 556 et seq., pp. 782, 814 et seq.) Any investigation of a potentially defamatory Internet posting is thus a daunting and expensive challenge. For that reason, we have observed that even when a defamation claim is “clearly nonmeritorious,” the threat of liability “ultimately chills the free exercise of expression.” (Baker v. Los Angeles Herald Examiner (1986) 42 Cal.3d 254, 268 [228 Cal.Rptr. 206, 721 P.2d 87]; see also Time, Inc. v. Hill (1967) 385 U.S. 374, 389 [17 L.Ed.2d 456, 87 S.Ct. 534].)

Nor are we convinced by the observation that a “distributor” faces no liability without notice. Distributors are liable not merely upon receiving notice from a third party, but also if they independently “knew or had reason to know” of the defamatory statement. (Osmond v. EWAP, Inc., supra, 153 Cal.App.3d 842, 854; Prosser & Keeton, The Law of Torts, supra, § 113, p. 811; Rest.2d Torts, § 581, subd. (1).) Thus, as the Zeran court pointed out, this aspect of distributor liability would discourage active monitoring of Internet postings. (Zeran, supra, 129 F.3d at p. 333.) It could also motivate providers to insulate themselves from receiving complaints. Such responses would frustrate the goal of self-regulation.

The third practical implication noted in Zeran is no less compelling, and went unaddressed by the Court of Appeal. Notice-based liability for service providers would allow complaining parties to impose substantial burdens on the freedom of Internet speech by lodging complaints whenever they were displeased by an online posting. (Zeran, supra, 129 F.3d at p. 333.) The volume and range of Internet communications make the “heckler’s veto” a real threat under the Court of Appeal’s holding. The United States Supreme Court has cautioned against reading the CDA to confer such a broad power of censorship on those offended by Internet speech. (Reno v. American Civil Liberties Union, supra, 521 U.S. at p. 880.)

The great variety of Internet publications, and the different levels of content control that may be exercised by service providers and users, do not undermine the conclusion that Congress intended to create a blanket immunity from tort liability for online republication of third party content. Requiring providers, users, and courts to account for the nuances of common law defamation, and all the various ways they might play out in the Internet environment, is a herculean assignment that we are reluctant to impose. We *58conclude the Zeran court accurately diagnosed the problems that would attend notice-based liability for service providers.

Finally, we cannot ignore another practical implication raised by Rosenthal and amicus curiae eBay Inc. Adopting a rule of liability under section 230 that diverges from the rule announced in Zeran and followed in all other jurisdictions would be an open invitation to forum shopping by defamation plaintiffs. (Cf. Webb v. Superior Court (1990) 225 Cal.App.3d 990, 1000 [275 Cal.Rptr. 581].) This consideration provides strong justification for following the approach we endorsed in Etcheverry v. Tri-Ag Service, Inc. (2000) 22 Cal.4th 316, 320-321 [93 Cal.Rptr.2d 36, 993 P.2d 366]: “While we are not bound by decisions of the lower federal courts, even on federal questions, they are persuasive and entitled to great weight. [Citation.] Where lower federal precedents are divided or lacking, state courts must necessarily make an independent determination of federal law [citation], but where the decisions of the lower federal courts on a federal question are ‘both numerous and consistent,’ we should hesitate to reject their authority [citation].”18

C. “User” Liability

The “distributor” liability theory endorsed by the Court of Appeal recognizes no distinction between Internet service providers and individuals. Individual Internet “users” like Rosenthal, however, are situated differently from institutional service providers with regard to some of the principal policy considerations discussed by the Zeran court and reflected in the Congressional Record. In particular, individuals do not face the massive volume of third party postings that providers encounter. Self-regulation is a far less challenging enterprise for them. Furthermore, service providers, no matter how active or passive a role they take in screening the content posted by users of their services, typically bear less responsibility for that content than do the users. Users are more likely than service providers to actively engage in malicious propagation of defamatory or other offensive material. These considerations bring into question the scope of the term “user” in section 230, and whether it matters if a user is engaged in active or passive conduct for purposes of the statutory immunity.

“User” is not defined in the statute, and the limited legislative record does not indicate why Congress included users as well as service providers under the umbrella of immunity granted by section 230(c)(1). The standard *59rules of statutory construction, however, yield an unambiguous result. We must begin with the language employed by Congress and the assumption that its ordinary meaning expresses the legislative purpose. (Engine Mfrs. Assn. v. South Coast Air Quality Management Dist. (2004) 541 U.S. 246, 252 [158 L.Ed.2d 529, 124 S.Ct. 1756]; see also Hassan v. Mercy American River Hospital (2003) 31 Cal.4th 709, 715 [3 Cal.Rptr.3d 623, 74 P.3d 726].) “User” plainly refers to someone who uses something, and the statutory context makes it clear that Congress simply meant someone who uses an interactive computer service.

Section 230(c)(1) refers directly to the “user of an interactive computer service.” Section 230(f)(2) defines “interactive computer service” as “any information service, system, or access software provider that provides or enables computer access by multiple users to a computer server, including specifically a service or system that provides access to the Internet . . . .” Section 230(a)(2) notes that such services “offer users a great degree of control over the information that they receive,” and section 230(b)(3) expresses Congress’s intent “to encourage the development of technologies which maximize user control over what information is received by individuals, families, and schools who use the Internet and other interactive computer services.” Thus, Congress consistently referred to “users” of interactive computer services, specifically including “individuals” in section 230(b)(3).

There is no reason to suppose that Congress attached a different meaning to the term “user” in section 230(c)(1). (See Gustafson v. Alloyd Co. (1995) 513 U.S. 561, 570 [131 L.Ed.2d 1, 115 S.Ct. 1061]; Hassan v. Mercy American River Hospital, supra, 31 Cal.4th at p. 716.) Rosenthal used the Internet to gain access to newsgroups where she posted Bolen’s article about Polevoy. She was therefore a “user” under the CDA, as the parties conceded below. Nor is there any basis for concluding that Congress intended to treat service providers and users differently when it declared that “[n]o provider or user of an interactive computer service shall be treated as [a] publisher or speaker . . . .” (§ 230(c)(1).) We cannot construe the statute so as to render the term “user” inoperative. (Duncan v. Walker (2001) 533 U.S. 167, 174 [150 L.Ed.2d 251, 121 S.Ct. 2120]; Hassan v. Mercy American River Hospital, supra, 31 Cal.4th at pp. 715-716.) We note that in cases where an individual’s role as operator of a Web site raised a question as to whether he was a “service provider” or a “user,” the courts found it unnecessary to resolve the issue because the statute confers immunity on both. (Batzel v. Smith, supra, 333 F.3d at p. 1030; Donato v. Moldow, supra, 865 A.2d at p. 719; see also Barrett v. Fonorow, supra, 799 N.E.2d at pp. 919, 922.)

Polevoy urges us to distinguish between “active” and “passive” Internet use, and to restrict the statutory term “user” to those who engage in passive *60use. He notes that subdivisions (a)(2) and (b)(3) of section 230 refer to information “received” by users. He also observes that the caption of subdivision (c) is “Protection for ‘Good Samaritan’ blocking and screening of offensive material.” From these premises, Polevoy reasons that the term “user” must be construed to refer only to those who receive offensive information, and those who screen and remove such information from an Internet site. He argues that those who actively post or republish information on the Internet are “information content providers” unprotected by the statutory immunity. “Information content provider” is defined as “any person or entity that is responsible, in whole or in part, for the creation or development of information provided through the Internet or any other interactive computer service.” (§ 230(f)(3).)

Polevoy’s view fails to account for the statutory provision at the center of our inquiry: the prohibition in section 230(c)(1) against treating any “user” as “the publisher or speaker of any information provided by another information content provider.” A user who merely receives information on a computer without making it available to anyone else would be neither a “publisher” nor a “speaker.” Congress obviously had a broader meaning in mind. Nor is it clear how a user who removes a posting may be deemed “passive” while one who merely allows a posting to remain online is “active.” Furthermore, Congress plainly did not intend to deprive all “information content providers” of immunity, because the reference to “another” such provider in section 230(c)(1) presumes that the immunized publisher or speaker is also an information content provider. (See Carafano v. Metrosplash.com, Inc., supra, 339 F.3d at p. 1125; Donato v. Moldow, supra, 865 A.2d at p. 720.)19

The distinction between “active” and “passive” use was explored in Batzel v. Smith, supra, 333 F.3d 1018. Smith sent an e-mail to the operator of a Web site devoted to museum security and stolen art, accusing Batzel of possessing paintings that may have been stolen by the Nazis during World War II. The operator posted the message on the Web site, with some changes, and distributed it to the subscribers of his e-mail newsletter. Batzel sued Smith and the operator for defamation. The trial court denied the operator’s motion to strike the complaint under the California anti-SLAPP statute (Code Civ. Proc., § 425.16). (Batzel, at pp. 1020-1023.)

*61The court of appeals vacated the order denying the motion, remanded, and directed the trial court to determine whether the operator should reasonably have known Smith intended his e-mail to be published on the Internet. If not, the court reasoned the message was not “provided” by another “information content provider” under section 230, and the operator would not be immune from liability.20 (Batzel v. Smith, supra, 333 F.3d at p. 1035.) The relevant discussion for our purposes arose from the dissent expressed in a concurring and dissenting opinion.

The Batzel dissent criticized the majority for adopting a rule that provides Internet intermediaries with immunity to spread information intended for republication, “licensing] professional rumor-mongers and gossip-hounds to spread false and hurtful information with impunity.” (Batzel v. Smith, supra, 333 F.3d at p. 1038 (conc. & dis. opn. of Gould, J.).) The dissent proposed a rule based on the defendant’s actions instead of the author’s intent. It would “hold that the CDA immunizes a defendant only when the defendant took no active role in selecting the questionable information for publication. If the defendant took an active role in selecting information for publication, the information is no longer ‘information provided by another’ within the meaning of § 230.” (Ibid.)

The dissent reasoned that information actively selected for republication has been “transformed . . . bolstered, [and] strengthened to do more harm if it is wrongful.” (Batzel v. Smith, supra, 333 F.3d at p. 1039 (cone. & dis. opn. of Gould, J.).) It acknowledged that service providers cannot be expected to screen the millions of messages sent over their networks for offensive content. However, it argued that a person who does actively screen communications to select some for republication is able to detect defamatory content and should not be immunized. The dissent would grant immunity to bulletin board moderators and the like if they did not actively select among messages for publication, but would expose them to liability if they made a conscious decision to disseminate a particular defamatory communication. Congress’s goal of encouraging self-regulation would be furthered, according to the dissent, because those who remove all or part of an offensive message would be immune. The dissenting justice did not believe Congress intended to immunize those who select defamatory information for distribution on the Internet. (Id. at pp. 1039-1040.)21

*62The Batzel majority responded that no logical distinction can be drawn between a defendant who actively selects information for publication and one who screens submitted material, removing offensive content. “The scope of the immunity cannot turn on whether the publisher approaches the selection process as one of inclusion or removal, as the difference is one of method or degree, not substance.” (Batzel v. Smith, supra, 333 F.3d at p. 1032.) We agree with this reasoning. Furthermore, we reject the dissent’s view that actively selected and republished information is no longer “information provided by another information content provider” under section 230(c)(1). All republications involve a “transformation” in some sense. A user who actively selects and posts material based on its content fits well within the traditional role of “publisher.” Congress has exempted that role from liability.

As Rosenthal points out, the congressional purpose of fostering free speech on the Internet supports the extension of section 230 immunity to active individual “users.” It is they who provide much of the “diversity of political discourse,” the pursuit of “opportunities for cultural development,” and the exploration of “myriad avenues for intellectual activity” that the statute was meant to protect. (§ 230(a)(3).) The approach taken by the Batzel dissent would tend to chill the free exercise of Internet expression, and could frustrate the goal of providing an incentive for self-regulation. A user who removed some offensive content might face liability for “actively selecting” the remaining material. Users in this position, no less than the service providers discussed by the Zerare court, would be motivated to delete marginally offensive material, restricting the scope of online discussion. Some users, at least those like Rosenthal who engage in high-volume Internet posting, might be discouraged from screening third party content. Although individual users may face the threat of liability less frequently than institutional service providers, their lack of comparable financial and legal resources makes that threat no less intimidating.

We conclude there is no basis for deriving a special meaning for the term “user” in section 230(c)(1), or any operative distinction between “active” and “passive” Internet use. By declaring that no “user” may be treated as a “publisher” of third party content, Congress has comprehensively immunized republication by individual Internet users.

D. Conclusion

We share the concerns of those who have expressed reservations about the Zeran court’s broad interpretation of section 230 immunity. The *63prospect of blanket immunity for those who intentionally redistribute defamatory statements on the Internet has disturbing implications. Nevertheless, by its terms section 230 exempts Internet intermediaries from defamation liability for republication. The statutory immunity serves to protect online freedom of expression and to encourage self-regulation, as Congress intended. Section 230 has been interpreted literally. It does not permit Internet service providers or users to be sued as “distributors,” nor does it expose “active users” to liability.

Plaintiffs are free under section 230 to pursue the originator of a defamatory Internet publication. Any further expansion of liability must await congressional action.

III. DISPOSITION

The judgment of the Court of Appeal is reversed.

George, C. J., Kennard, J., Baxter, J., Werdegar, J., Chin, J., and Moreno, J., concurred.