This defamation case concerns Mind, Body and Soul, a book written by Scott Weier. In the author’s words, the book is “based on my life.” It discusses Scott’s *440personal transformation, largely through his relationship with God, following his divorce “on bad terms” from his first wife. Scott’s ex-wife and her father concluded the book falsely accused them of lying, abuse, and suffering from mental illness. They sued Scott and Author Solutions, Inc. (ASI), the company that produced the book, for libel, invasion of privacy, and intentional infliction of emotional distress. The district court, finding issues of fact, denied both defendants’ motions for summary judgment. The author and the publisher appealed.
On our review, we uphold the denial of Scott’s motion for summary judgment for substantially the reasons set forth in the district court’s thorough opinion. However, we hold that ASI, as a bona fide book publisher, should be considered a “media defendant.” Therefore, we find that ASI was entitled to summary judgment because plaintiffs failed to provide sufficient proof to establish a prima facie case under the established standards applicable to such defendants. We also decline Scott and ASI’s invitation to revise our common law of defamation at this time.
I. Background Facts and Proceedings.
After a contentious divorce which apparently resulted in a severing of Scott’s ties with his daughters as well as his ex-wife, Scott wrote a 253-page memoir entitled Mind, Body and Soul.1 In it he described his own shortcomings and the development of his personal relationship with God and offered advice for others on their own spiritual journeys. He also criticized his ex-wife, Beth Weier, in various respects— accusing her of lying to their daughters, being a bad parent, having a lack of religious conviction, and generally being mean and spiteful. In a key passage, he alleged that Beth’s father, Gail Bierman, molested her as a child and that she suffered from either bipolar disorder or borderline personality disorder as a result. In one sentence summarizing a theme of the book, Scott wrote, “Satan (through my ex) set out to destroy my life, and has upended it completely, but through doing so, it has completely changed my life for the good!”
In late 2008, Scott enlisted the services of ASI to publish his book. For a total fee of $3183.81, ASI formatted and typeset the manuscript, designed the cover, and provided 250 copies of the book for Scott to self-distribute to local bookstores, friends, and family.
ASI offered proofreading and editing services, but Scott declined to purchase them. ASI did run a simple software program on the text that it described as a “manuscript scrub.” This program is a macro in Microsoft Word designed to identify passages that contain certain “buzz” words that might have the potential for being problematic. As an ASI employee explained:
What happens is all these words show up, and they’re from all sorts of topics. Could be trademark issues or copyright issues. It’s a variety of things. It can also be for offensive language. There’s other things I look for as well just to make sure that there isn’t some sort of hate literature or something like that.
So what will happen is the file will be created, the scrub will be created and all these words will be highlighted.... So it’s just a broad range of things you try to go through and kind of size it up as quickly as possible.
After performing the manuscript scrub on Scott’s book, ASI’s employee received a *441highlighted printout that he “speed read.” In one identified passage, Scott had written, “I was molested by an uncle sexually as a young child.” The employee did not do anything about this passage after verifying with Scott that he had more than one uncle. The same employee also initially noted the following passage:
The two women we spoke of earlier, they were both molested by their fathers, or at least that is what they told me. And both of them were bipolar or borderline personality disorder, which is a fairly normal result of this type of sin against a child. Why does one person end up with mental issues and the other does not?
ASI’s employee did not take action on this passage because he “didn’t think there was enough information about the women.” Thus, ASI did not require any changes to the book prior to publication.
ASI did not promote the book but did provide guidance and tips to Scott on how to market his book himself. Scott distributed twenty to thirty of his 250 copies of the book to friends, family, and businesses in the Clear Lake area. The rest of the books remain stored in his parents’ basement. ASI also offered the book on its own website, where three copies were sold, and through Amazon.com, where one copy was sold.
Following the book’s release, Beth learned from a friend that Scott had written it and had made reference to her in it. She obtained a copy in February 2009, read it, and discovered various references to her and her daughters, including passages that appeared to indicate Beth had been abused by her father and suffered from bipolar or borderline personality disorder. She believed those statements, as well as others in the book, were false and defamatory. She and her father retained counsel and sent a cease and desist letter to Scott and ASI. Neither Scott nor ASI took action in response to the letter.
On February 24, 2009, Beth and Gail filed a petition in the Polk County District Court alleging libel per se, false light invasion of privacy, and intentional infliction of emotional distress. The petition specifically identified thirty-two excerpts from the book as being defamatory.2 The plaintiffs also sought and were granted a temporary injunction to prevent further distribution of the book during the pendency of the litigation.3
Scott and ASI filed separate answers to Beth and Gail’s petition.4 Later, after dis*442covery, all parties filed motions for summary judgment. Beth and Gail sought partial summary judgment on their libel claims, requesting an adjudication that certain passages in the book were libelous per se.
ASI filed a motion for summary judgment urging dismissal of all claims against it. ASI contended that most of Beth and Gail’s libel claims should fail as a matter of law because the statements identified by the plaintiffs either were admitted to be true, were not about the plaintiffs, were not provably false, or were not defamatory. Additionally, ASI argued that Beth and Gail could not establish the elements of libel and were not entitled to presumptions under a libel per se theory because ASI was a media defendant. ASI’s motion further argued that the claims for false light and intentional infliction of emotional distress should be summarily dismissed because they were simply libel claims under a different label. Alternatively, ASI maintained that the plaintiffs could not establish the publicity or fault requirements of their false light claims, and the plaintiffs could not establish the necessary elements of intentional infliction of emotional distress. ASI also sought summary judgment on the plaintiffs’ request for punitive damages against it.
Scott’s motion for summary judgment advanced most of the same arguments as ASI’s motion, although he did not contend that he was a media defendant.
On September 15, 2010, the district court issued a twenty-three-page ruling on the parties’ motions. The court concluded the statements in Scott’s book regarding Gail’s alleged abuse of Beth and Beth’s resulting mental illness constituted libel per se and granted Beth and Gail’s motion for partial summary judgment on that ground. The court then turned to whether ASI was a media defendant. It found it was not:
ASI is not the New York Times, or any other media entity. Rather it is a business which contracts to publish documents for private authors. And while its authors may, in some instances, have first amendment rights, the rights retained by ASI have nothing to do with the First Amendment.... Based on the Court[’]s earlier determination that certain statements in the Book are libelous per se, ASI should be treated here as any other private defendant would be in a libel per se action. Accordingly, the elements of falsity, malice, and damage can be presumed as to ASI and the only element the Plaintiffs would have to prove is publication.
The district court further concluded that even if the plaintiffs were required to prove the four elements of a libel claim, they had demonstrated a fact issue as to each element sufficient to survive summary judgment.
The district court also denied ASI’s motion for summary judgment on the false light invasion of privacy claim, finding fact issues existed as to whether ASI gave publicity to the book and whether ASI acted recklessly or with knowledge the book was false. The court, however, granted summary judgment to ASI on the intentional infliction of emotional distress claim, concluding the plaintiffs had failed to show a fact question as to the claimed outrageousness of ASI’s conduct. The district court also denied ASI’s motion for summary judgment on the question of punitive damages, finding a fact question existed as to whether ASI acted recklessly when it published the allegedly defamatory *443statements in the book. In addition, the district court denied Scott’s motion for summary judgment on all claims.
ASI and Scott applied for leave with this court to pursue an interlocutory appeal. We granted their applications and stayed proceedings pending appeal. We initially heard argument in this case last term. However, we then decided to hold this case over for reargument and further consideration in the current term. At that time, we asked the parties to provide supplemental briefing on whether this court should continue to recognize the doctrine of libel per se.
II. Scope of Review.
Our review of rulings on motions for summary judgment is for correction of errors at law. Kiesau v. Bantz, 686 N.W.2d 164, 171 (Iowa 2004). Summary judgment is appropriately granted when there is no genuine issue of material fact and the moving party is entitled to judgment as a matter of law. Id. When considering a motion for summary judgment, the record must be viewed in the light most favorable to the nonmoving party. Id.
We have explained that summary judgment “is afforded a unique role in defamation cases. Judges have a responsibility to determine whether allowing a case to go to a jury would ... endanger first amendment freedoms.” Jones v. Palmer Commc’ns, Inc., 440 N.W.2d 884, 889 (Iowa 1989) (citation and internal quotation marks omitted), overruled in part on other grounds by Schlegel v. Ottumwa Courier, 585 N.W.2d 217, 224 (Iowa 1998).
III. Discussion.
Defamation law in Iowa is a blend of three things: common law, some statutes set forth in chapter 659 of the Iowa Code, and First Amendment principles established by decisions of the United States Supreme Court. In its constitutional decisions, that Court has seemingly cleared a path for traditional common law defamation claims to proceed when the plaintiff is a private figure and the defamation concerns private matters. Less clear is whether the identity of the defendant as a media defendant changes the constitutional analysis.
Nonetheless, since the United States Supreme Court constitutionalized the law of defamation, our court has consistently viewed media defendant status as significant. When the defendant is a media defendant, we have said that presumptions of fault, falsity, and damages are not permissible, and thus the common law doctrine of libel per se cannot apply. We must now decide whether we should continue to recognize libel per se and the distinction between media and nonmedia defendants, and if so, where ASI belongs.
To frame this discussion, it is useful to review how we got to where we are today. Accordingly, we will first discuss Iowa’s historical law of defamation and then the United States Supreme Court’s landmark defamation eases, followed by our own response to those decisions. Against that backdrop, we will consider the defendants’ arguments that we should abandon libel per se in light of certain constitutional developments as well as the growth of the Internet.
A. Iowa’s Common Law of Defamation. Before 1964, “defamation law consisted primarily of a complex set of common-law rules developed by the state courts.” Jones, 440 N.W.2d at 890. At common law, defamation involved the following elements: (1) publication, (2) of a defamatory statement, (3) which was false and (4) malicious, (5) made of and concerning the plaintiff, (6) which caused injury. See Johnson v. Nickerson, 542 N.W.2d *444506, 510 (Iowa 1996); Vojak v. Jensen, 161 N.W.2d 100, 104, 108 (Iowa 1968), abrogated in part on other grounds by Barreca v. Nicholas, 683 N.W.2d 111, 119-21 (Iowa 2004). “Defamation includes the twin torts of libel and slander. Libel involves written statements, while slander involves oral statements.” Kiesau, 686 N.W.2d at 174 (Iowa 2004) (citation omitted). We recognized two types of libel: libel per se and libel per quod. Id. at 175.
Certain statements were held to be libelous per se, which meant they were “actionable in and of themselves without proof of malice, falsity or damage.” Vojak, 161 N.W.2d at 104. This was “based on the very nature of the language used.” Nickerson, 542 N.W.2d at 510. Libel per se statements have “ ‘a natural tendency to provoke the plaintiff to wrath or expose him to public hatred, contempt, or ridicule, or to deprive him of the benefit of public confidence or social intercourse.’ ” Id. (quoting Prewitt v. Wilson, 128 Iowa 198, 202, 103 N.W. 365, 367 (1905)). For example, “[i]t is libel per se to make published statements accusing a person of being a liar, a cheater, or thief.” Spencer v. Spencer, 479 N.W.2d 293, 296 (Iowa 1991). “To accuse a person of an indictable crime is defamation per se.” Huegerich v. IBP, Inc., 547 N.W.2d 216, 221 (Iowa 1996); see also Patrick J. McNulty, The Law of Defamation: A Primer for the Iowa Practitioner, 44 Drake L.Rev. 639, 648 (1996) [hereafter McNulty] (listing additional examples of libel per se).
In libel per quod cases, by contrast, a plaintiff must ordinarily prove all the above six elements, including “some sort of cognizable injury, such as injury to reputation.” Nickerson, 542 N.W.2d at 513; see also Suntken v. Den Ouden, 548 N.W.2d 164, 167 (Iowa Ct.App.1996). Further, “[h]urt feelings alone cannot serve as the basis of a defamation action.” Nickerson, 542 N.W.2d at 513. A statement was considered libelous per quod at common law if it was “necessary to refer to facts or circumstances beyond the words actually used to establish the defamation.” Id. at 510. Thus, a statement would be deemed libel per quod where the words in themselves were not considered sufficiently harmful to the plaintiff without further context. See, e.g., Ragland v. Household Fin. Corp., 254 Iowa 976, 982-83, 119 N.W.2d 788, 792 (Iowa 1963) (holding a statement that the plaintiff had not paid a debt was not libelous per se).
“Although [the per se] presumptions were attacked through the years, sometimes scornfully, they remained viable until the United States Supreme Court began to intervene in 1964.” McNulty, 44 Drake L.Rev. at 643-44 (footnote omitted).
B. The United States Supreme Court Intervenes. In 1964, the United States Supreme Court for the first time placed First Amendment boundaries on the common law of defamation. In New York Times v. Sullivan, 376 U.S. 254, 279-80, 84 S.Ct. 710, 726, 11 L.Ed.2d 686, 706 (1964), it overturned a libel judgment that an Alabama city commissioner had obtained against the New York Times and announced that “[t]he constitutional guarantees require ... a federal rule that prohibits a public official from recovering damages for a defamatory falsehood relating to his official conduct unless he proves that the statement was made with ‘actual malice’ — that is, with knowledge that it was false or with reckless disregard of whether it was false or not.” Seven years later, in Rosenbloom v. Metromedia, Inc., 403 U.S. 29, 43-44, 91 S.Ct. 1811, 1819-20, 29 L.Ed.2d 296, 312 (1971), a plurality of the Court extended this protection to private persons when the defamatory statements concerned matters of general or public interest.
*445However, just three years after that, the Court changed course somewhat in a case brought by an attorney who was neither a public official nor a public figure. Gertz v. Robert Welch, Inc., 418 U.S. 323, 325-32, 94 S.Ct. 2997, 3000-03, 41 L.Ed.2d 789, 797-801 (1974). Although the magazine article in question involved a matter of public interest, the Court acknowledged the criticism of the Rosenbloom dissent that providing special protection for speech on matters of public concern “would involve the courts in the dangerous business of deciding ‘what information is relevant to self-government.’ ” Id. at 339, 94 S.Ct. at 3006-07, 41 L.Ed.2d at 805 (quoting Rosenbloom, 403 U.S. at 79, 91 S.Ct. at 1837, 29 L.Ed.2d at 332 (Marshall, J., dissenting)). Thus, in Gertz, the Court held “that, so long as they do not impose liability without fault, the States may define for themselves the appropriate standard of liability for a publisher or broadcaster of defamatory falsehood injurious to a private individual.” Id. at 347, 94 S.Ct. at 3010, 41 L.Ed.2d at 809. Gertz also concluded that “the States may not permit recovery of presumed or punitive damages, at least when liability is not based on a showing of knowledge of falsity or reckless disregard for the truth.” Id. at 349, 94 S.Ct. at 3011, 41 L.Ed.2d at 810.
The defendant in Gertz was a magazine publisher, id. at 325, 94 S.Ct. at 3000, 41 L.Ed.2d at 797, and the Supreme Court’s opinion included extensive references to newspapers, broadcasters, publishers, and news media, see id. at 340, 94 S.Ct. at 3007, 41 L.Ed.2d at 805-06. Accordingly, the decision has often been interpreted as distinguishing between media and nonme-dia defendants with its fault and damage proof requirements applying to lawsuits involving a media defendant. See McNulty, 44 Drake L.Rev. at 695 n. 574.
However, a decade after the Gertz decision was filed, the Court added another layer of complexity in Dun & Bradstreet, Inc. v. Greenmoss Builders, Inc., 472 U.S. 749, 105 S.Ct. 2939, 86 L.Ed.2d 593 (1985). In another plurality opinion, the court re-injected considerations of whether the challenged publication related to a matter of public concern. The Court’s plurality opinion interpreted the holding in Gertz as limited to matters of public concern, while concluding the First Amendment imposed no restrictions on speech of purely private concern about a private party plaintiff. Dun & Bradstreet, 472 U.S. at 763, 105 S.Ct. at 2947, 86 L.Ed.2d at 605 (plurality opinion). Thus, in Dun & Bradstreet, the Court upheld a state supreme court decision reinstating a jury verdict that awarded presumed and punitive damages to a business defamed by a false credit report without proof of malice. Id.
Two years later, the Supreme Court again addressed the intersection of the First Amendment and libel claims in Philadelphia Newspapers Inc. v. Hepps, 475 U.S. 767, 106 S.Ct. 1558, 89 L.Ed.2d 783 (1986). In Hepps, the plurality concluded a private figure plaintiff must bear the burden of proving the falsity of speech against a media defendant when the speech is of public concern. Hepps, 475 U.S. at 776-77, 106 S.Ct. at 1563-64, 89 L.Ed.2d at 793 (plurality opinion). The Court added that it did not need to “consider what standards would apply if the plaintiff sues a nonmedia defendant.” Id. at 779 n. 4, 106 S.Ct. at 1565 n. 4, 89 L.Ed.2d at 794 n. 4. But Justice Brennan and Justice Blackmun, whose votes were necessary to form the majority, indicated in a separate concurring opinion that they would not accept a media/nonmedia distinction. See id. at 779-80, 106 S.Ct. at 1565-66, 89 L.Ed.2d at 795 (Brennan, J., concurring).
*446C. Our Court Responds to the United States Supreme Court’s Decisions and Embraces the Media Defendant/Nonmedia Defendant Distinction. We applied the New York Times v. Sullivan framework in a few early cases involving public officials or public figures. See Anderson v. Low Rent Hous. Comm’n, 304 N.W.2d 239, 248-49 (Iowa 1981); Blessum v. Howard Cnty. Bd. of Supervisors, 295 N.W.2d 836, 843 (Iowa 1980); McCarney v. Des Moines Register & Tribune Co., 239 N.W.2d 152, 156 (Iowa 1976). Then, after the Supreme Court decided Gertz, but before it decided Dun & Bradstreet, we addressed Gertz’s application to Iowa libel law in a 1984 case involving a private figure. See Vinson v. Linn-Mar Cmty. Sch. Dist., 360 N.W.2d 108, 117-18 (Iowa 1984). In Vinson, a former school bus driver brought various claims against a school district including a claim of defamation. Id. at 111. The defendant school system argued that we should discard existing Iowa law regarding libel per se and “adopt the standard mandated for actions by private individuals against media defendants in Gertz.” Id. at 117. We noted that Vinson involved a nonmedia defendant and found the speech was not constitutionally protected under Gertz. We concluded:
In this situation where only a private plaintiff and non-media defendant are involved, the common law standard does not threaten the free and robust debate of public issues or a meaningful dialogue about self-government, or freedom of the press. We refuse to extend the Gertz holding to actions between a private individual and a non-media defendant.
Id. at 118. Vinson remains the law in Iowa to this day.5
In 1989, in Jones, we were confronted with a libel case against a media defendant. The case involved a television story discussing the termination of a black firefighter’s employment after he had failed a written examination. Jones, 440 N.W.2d at 889-90. The firefighter had been hired as part of a federal court consent decree entered in an employment discrimination class action case. Id. at 889. The firefighter sued the owner of the television channel. Id. at 888. The media defendant asked us to require that the plaintiff prove actual malice. Id. at 896. We again read Gertz as not permitting liability without fault to be imposed on a media defendant. Id. However, we decided to adopt a negligence standard, rather than an actual malice standard. Id. at 896-99. We did so even though the media defendant argued that the television report had been on a matter of public concern. Id. at 897-98.
In rejecting an actual malice standard, we relied in part upon the language of our own constitution which provides “[ejvery person may speak, write, and publish his sentiments on all subjects, being responsible for the abuse of that right.” Id. at 898 (quoting Iowa Const, art. I, § 7 (emphasis added)). “[T]his express concern for the abuse of free speech is not found in the United States Constitution.” Id. We not*447ed, “Several other states have interpreted similar clauses in state constitutions to justify the adoption of a negligence standard for private plaintiffs in a defamation action.” Id.
Seven years later, in Nickerson, which again involved a media defendant, we reiterated our view of the importance of the media defendant/nonmedia defendant distinction. In that case, the foreman of a jury that found an African-American defendant guilty of murder filed a defamation action after the Des Moines Register printed a story alleging links between the foreman and a white supremacist group. Nickerson, 542 N.W.2d at 509. We took the position that under Gertz, a private party must establish fault — and for that matter actual damages — to bring a case against a media defendant. Id. at 511. “Hence, in cases against a media defendant, the distinction between libel and libel per se has become irrelevant.” Id. We summarized:
[T]o establish a prima facie defamation action against a media defendant, a private figure plaintiff must prove (1) publication (2) of a defamatory statement (3) concerning the plaintiff (4) in a negligent breach of the professional standard of care (5) that resulted in demonstrable injury.
Id.
Johnson explained that to prevail in a defamation action against a media defendant, a plaintiff must “prove some sort of cognizable injury, such as injury to reputation. Hurt feelings alone cannot serve as the basis of a defamation action.” Id. at 513 (citation omitted). We indicated that when suing media defendants for defamation, plaintiffs no longer benefit from presumed fault or damages. Id. We also noted that “[b]oth public officials and private individuals must prove the falsity of the challenged statements.” Id. at 511 n. 3 (citing Hepps, 475 U.S. at 775-76, 106 S.Ct. at 1563-64, 89 L.Ed.2d at 792).
Two years later, in 1998, we examined more closely the damages that must be proved by a private plaintiff in a libel action against a media defendant. Schlegel, 585 N.W.2d at 222-23 (Iowa 1998). In that case, the Ottumwa Courier incorrectly reported that a local lawyer had declared bankruptcy, and the lawyer sued for defamation. Id. at 220. We reaffirmed that the libel per se damage presumption does not apply when the defendant is a member of the media; a plaintiff needs to prove actual damages. Id. at 222-23. We acknowledged that Gertz permitted a private plaintiff to recover against a media defendant under a broad formulation of actual damages which included humiliation and mental anguish. Id. at 223-24. We concluded, however, that an Iowa plaintiff must establish actual reputational harm when suing a media defendant, and not merely emotional distress or humiliation, before he or she may recover for any parasitic damages such as personal humiliation or mental anguish. Id.
Three years after that, in Caveman Adventures UN, Ltd. v. Press-Citizen Co., 633 N.W.2d 757 (Iowa 2001), abrogated in part on other grounds by Barreca, 683 N.W.2d at 119-21, we addressed the standards for awarding punitive damages against media defendants. There, an electronics store paid a newspaper to run an advertisement making unflattering claims about a competing store. Caveman Adventures, 633 N.W.2d at 760. The competitor sued the newspaper. Id. We reiterated that “[i]n the wake of Gertz, the common-law rules of libel recovery have been most altered with regard to private-party actions against media publishers or broadcasters.” Id. at 761. We held the plaintiff could not recover punitive damages from the newspaper absent a show*448ing that the newspaper had acted with actual malice; because the plaintiff had failed to make that showing, the judgment had to be reversed. Id. at 762.
Throughout the above cases, it appears we have relied largely on a reading of Gertz that gives legal effect to defendants’ media status. We have also interpreted our common law, outside the libel per se context, to place higher fault and damage burdens on plaintiffs. As a result, in private plaintiffiprivate interest cases, media status is highly determinative in Iowa. A media defendant benefits from the bar on presumed damages and the requirement to prove fault and falsity, whereas a nonme-dia defendant is subject to presumptions of damages, falsity, and malice if a traditional case of defamation per se has been established.
While a close reading of the United States Supreme Court cases on point reveals that we may not have been compelled to arrive at this distinction, the media/non-media dichotomy is nonetheless a well-established component of Iowa’s defamation law. As a result, libel per se is available only when a private figure plaintiff sues a nonmedia defendant for certain kinds of defamatory statements that do not concern a matter of public importance. See, e.g., Kiesau, 686 N.W.2d at 175. In these cases, if the alleged defamatory statements have “a natural tendency to provoke the plaintiff to wrath or expose him to public hatred, contempt, or ridicule,” the plaintiff need not prove that the statement actually damaged her or him; damages are presumed. Nickerson, 542 N.W.2d at 510 (citation and internal quotation marks omitted).
D. Should We Abandon Libel Per Se? ASI and Scott ask us to abandon what remains of libel per se in Iowa and require that defamation plaintiffs always prove falsity, fault, and damages to reputation. They contend we should abolish the twenty-eight-year-old distinction in our caselaw between media and nonmedia defendants and, effectively, the doctrine of libel per se dating back to the nineteenth century. They maintain we should do so because the distinction is both unconstitutional and unwise as a matter of public policy. In ASI and Scott’s view, it is impermissible under the First Amendment and article I, section 7 of the Iowa Constitution for state common law to afford more protection to media defendants than to other defendants. They also insist that the presumptions associated with libel per se have outlived their usefulness, and that technological developments — specifically the rise of the Internet and electronic publishing — have rendered the media/nonmedia defendant distinction obsolete. We now turn to these arguments.
1. Federal constitutional considerations. Notwithstanding ASI and Scott’s contentions, we are not persuaded that our current libel law in Iowa transgresses First Amendment boundaries. The United States Supreme Court has never invalidated the common law libel presumptions as they now apply in Iowa— to private plaintiffiprivate concern cases against nonmedia defendants. Most persuasive on this point, Dun & Bradstreet actually approved the libel per se presumption of damages in a private plaintiff/private concern case against a nonme-dia defendant. The facts and result of that case are instructive here. In that case, a construction contractor sued a credit reporting company for defamation when the company falsely reported that the contractor had filed for bankruptcy. Dun & Bradstreet, 472 U.S. at 751-52, 105 S.Ct. at 2941, 86 L.Ed.2d at 597-98 (plurality opinion). The contractor was a private plaintiff, the credit report was not a matter of public concern, and the reporting com*449pany was a nonmedia defendant. Id. A total of five justices agreed the First Amendment allowed common law presumed and punitive damages under those facts. Id. at 768, 105 S.Ct. at 2947, 86 L.Ed.2d at 605 (plurality opinion); id. at 764, 105 S.Ct. at 2948, 86 L.Ed.2d at 605-06 (Burger, C.J., concurring); id. at 774, 105 S.Ct. at 2953, 86 L.Ed.2d at 612 (White, J., concurring).
The two justices who concurred in the judgment went further, arguing that traditional common law libel rules could be applied whenever the plaintiffs were private citizens. Id. at 763-64, 105 S.Ct. at 2948, 86 L.Ed.2d at 605 (Burger, C.J., concurring) (“I preferred to allow this area of law to continue to evolve as it had up to then with respect to private citizens rather than embark on a new doctrinal theory....” (Alterations omitted.) (Citation and internal quotation marks omitted.)); id. at 767, 105 S.Ct. at 2949, 86 L.Ed.2d at 607-08 (White, J., concurring) (“[C]ommon-law remedies should be retained for private plaintiffs”).
In Iowa, the unaltered common law per se rule applies only on facts like those in Dun & Bradstreet — private plaintiff, private concern, and nonmedia defendant. That is, it is consistent with the result in Dun & Bradstreet. No subsequent Supreme Court decision has held otherwise on facts like those before us.
Some observers, as well as the four Dun & Bradstreet dissenters, note that in that case six Justices declined to draw a First Amendment line based on defendants’ media status (i.e., the four dissenters plus the two who concurred in the judgment). See, e.g., id. at 783-84, 105 S.Ct. at 2958, 86 L.Ed.2d at 618-19 (Brennan, J., dissenting). But this does not render Iowa’s decision to honor such a distinction unconstitutional. The two justices who concurred in the Dun & Bradstreet judgment would have allowed libel per se claims by private plaintiffs to proceed against both media and nonmedia defendants. Id. at 763-64, 105 S.Ct. at 2947-48, 86 L.Ed.2d at 605-06 (Burger, C.J., concurring); id. at 772-74, 105 S.Ct. at 2952-53, 86 L.Ed.2d at 611-12 (White, J., concurring). Iowa can make its defamation law more protective of defendants than the First Amendment requires. And less than a year after Dun & Bradstreet, a plurality of the Court reemphasized defendants’ media status in Repps, and specifically withheld judgment on whether Gertz restrictions apply to nonmedia defendants. 475 U.S. at 777, 779 n. 4,106 S.Ct. at 1564, 1565 n. 4, 89 L.Ed.2d at 793, 797 n. 4 (plurality opinion).
ASI and Scott also argue a recent United States Supreme Court decision, United States v. Alvarez, — U.S. -, 132 S.Ct. 2537, 183 L.Ed.2d 574 (2012), has implicitly invalidated libel per se presumptions by recognizing a First Amendment right to make factually false statements. That case invalidated the Stolen Valor Act, a law making it a crime to falsely claim receipt of a military decoration or medal authorized by Congress. Alvarez, — U.S. at -, 132 S.Ct. at 2543, 183 L.Ed.2d at 585-86 (plurality opinion). The four justices in the plurality concluded that, without some legally cognizable harm, the falsity of defendant’s speech was not enough to justify government penalties. Id. at -, 132 S.Ct. at 2545-46, 183 L.Ed.2d at 588-89. Further, the law did not implicate one of the “few historic and traditional categories” where “content-based restrictions on speech have been permitted.” Id. at -, 132 S.Ct. at 2544, 183 L.Ed.2d at 586-87 (citation and internal quotation marks omitted). Those traditionally unprotected categories, the plurality noted, included defamation. Id.
*450The two justices who concurred in the judgment in Alvarez applied a different level of scrutiny to the Stolen Valor Act, but also observed that laws punishing false fact statements were permissible when they “limit the scope of their application, sometimes by requiring proof of specific harm to identifiable victims; sometimes by specifying that the lies be made in contexts in which a tangible harm to others is especially likely to occur; and sometimes by limiting the prohibited lies to those that are particularly likely to produce harm.” Id. at -, 132 S.Ct. at 2554, 183 L.Ed.2d at 597 (Breyer, J., concurring). And three dissenting justices would have upheld the constitutionality of the Stolen Valor Act. Id. at -, 132 S.Ct. at 2556-65, 183 L.Ed.2d at 600-10 (Alito, J., dissenting).
ASI and Scott contend the libel per se presumptions cross the same constitutional threshold the Stolen Valor Act crossed, by imposing strict liability on nonmedia defendants for publishing false statements. The problem with this argument, however, is that both opinions making up the Alvarez majority specifically highlighted defamation as a traditional area where the law was constitutional because it did not punish statements merely because of their falsity. Id. at -, 132 S.Ct. at 2545, 183 L.Ed.2d at 587 (plurality opinion); id. at -, 132 S.Ct. at 2554, 183 L.Ed.2d at 597 (Breyer, J., concurring). “Defamation statutes focus upon statements of a kind that harm the reputation of another or deter third parties from association or dealing with the victim.” Id. at -, 132 S.Ct. at 2554, 183 L.Ed.2d at 597 (Breyer, J., concurring) (emphasis added).
It is noteworthy that both the plurality opinion and Justice Breyer’s concurrence not only recognize the continued vitality of defamation law, but also cite Gertz with approval. Id. at -, 132 S.Ct. at 2544, 183 L.Ed.2d at 587 (plurality opinion); id. at -, 132 S.Ct. at 2553, 183 L.Ed.2d at 597 (Breyer, J., concurring). As we have pointed out, our court has relied upon Gertz in large part to sustain its distinction between media and nonmedia defendants. See Caveman Adventures, 633 N.W.2d at 761-62; Schlegel, 585 N.W.2d at 224-26; Nickerson, 542 N.W.2d at 510-12; Jones, 440 N.W.2d at 896-97. In sum, Alvarez characterizes defamation as a category of false speech the government is allowed to restrict, so it is unlikely that case does anything to change the Court’s libel jurisprudence.
ASI also argues the Citizens United decision, which overturned federal prohibitions on election spending by corporations, has washed away any distinction between media and nonmedia defendants in libel actions. See Citizens United v. Fed. Election Comm’n, 558 U.S. 310, 130 S.Ct. 876, 175 L.Ed.2d 753 (2010). In justifying its position, the Citizens United majority noted that although the Bipartisan Campaign Reform Act did not apply to media corporations, one of the rationales asserted by the government in defending the Act could be used to restrict political speech by media corporations. Id. at -, 130 S.Ct. at 905-06, 175 L.Ed.2d at 790-91. The Court then elaborated:
The media exemption discloses further difficulties with the law now under consideration. There is no precedent supporting laws that attempt to distinguish between corporations which are deemed to be exempt as media corporations and those which are not. “We have consistently rejected the proposition that the institutional press has any constitutional privilege beyond that of other speakers.” With the advent of the Internet and the decline of print and broadcast media, moreover, the line between the media and others who wish to *451comment on political and social issues becomes far more blurred.
Id. at -, 180 S.Ct. at 905-06, 175 L.Ed.2d at 790 (citations omitted).
None of the Court’s discussion, however, addressed the law of defamation. This topic was touched on two years later in Alvarez, and as noted, the Court implicitly approved its earlier precedents. Nor did Citizens United suggest that anything prevents a state from affording more protection to media defendants in libel cases (whether they are corporations or not) than the federal constitutional minimum. In short, Citizens United, like Dun & Bradstreet before it, may indicate that a majority of the Court questions the constitutional significance of a media/nonmedia distinction. But it is impossible to find a hint in Citizens United or any other United States Supreme Court decision that states may not continue to recognize libel per se in private plaintiff/private concern/nonmedia defendant cases if they choose to do so.
2. Iowa constitutional considerations. We are likewise convinced that Iowa’s Constitution does not bar the application of libel per se to private plaintiff/private concern cases against nonmedia defendants. Article I, section 7 of our constitution provides:
Every person may speak, write, and publish his sentiments on all subjects, being responsible for the abuse of that right. No law shall be passed to restrain or abridge the liberty of speech, or of the press. In all prosecutions or indictments for libel, the truth may be given in evidence to the jury, and if it appears to the jury that the matter charged as libellous was true, and was published with good motives and for justifiable ends, the party shall be acquitted.
Iowa Const. art. I, § 7. We have said that “the Iowa Constitution generally imposes the same restrictions on the regulation of speech as does the federal constitution.” State v. Milner, 571 N.W.2d 7, 12 (Iowa 1997); see also In re Adoption of S.J.D., 641 N.W.2d 794, 802 (Iowa 2002).
In any event, to the extent there are textual differences between the First Amendment and article I, section 7, they do not support the elimination of libel per se. In the third sentence of section 7, our constitution’s framers specifically allowed for criminal libel, while providing that the defendant shall be acquitted if truth of the statement, “good motives” and “justifiable ends” were shown. This sentence mirrors the relevant text of the criminal libel statute that was in effect when our 1857 Constitution was adopted. See Iowa Code § 2769 (1851) (“In all prosecutions or indictments for libel the truth thereof may be given in evidence to the jury, and if it appear to them that the matter charged as libelous was true and was published with good motives and for justifiable ends the defendant shall be acquitted.”); see also id. § 2767 (“A libel is the malicious defamation of a person made public by any printing, writing, sign, picture, representation, or effigy, tending to provoke him to wrath or expose him to public hatred, contempt, or ridicule, or to deprive him of the benefits of public confidence and social intercourse.... ”). Thus, our constitutional framers clearly contemplated criminal libel — in fact, there was an existing criminal libel statute — and made provisions for it in the Iowa Constitution.
As ASI expounds in its supplemental brief, libel per se is a civil doctrine that derives from criminal libel. See Mosnat v. Snyder, 105 Iowa 500, 504, 75 N.W. 356, 358 (1898). Both doctrines impose a form of strict liability subject to a defense. If our constitution was written to expressly *452allow for criminal libel, it is difficult to see why it would not tolerate libel per se as well. The Iowa Constitution appears to recognize rather than undercut its validity.6
Also, unlike the First Amendment, article I, section 7 contains an “abuse clause.” See Iowa Const. art. I, § 7 (“being responsible for the abuse of that right”). In Jones, we found the abuse clause shows an “express concern for injury to reputation found in the Iowa Constitution.” Jones, 440 N.W.2d at 898. While recognizing that United States Supreme Court precedent did not allow liability without fault in private plaintiff/public concern/media defendant cases, we relied on the “abuse” clause to hold that no more than proof of negligence (rather than proof of malice) should be required. Id. at 898-99.
A number of other states have similar abuse language in the free speech clauses of their constitutions. Some, like Jones, hold that an abuse clause justifies a lower scienter requirement in libel cases where some proof of fault is required. See Traman v. Wood, 62 Ill.2d 184, 340 N.E.2d 292, 297 (1975) (“The freedom of speech provisions of both our former and present constitutions ... recognize the interest of the individual in the protection of his reputation, for they provide that the exercise of the right to speak freely shall not relieve the speaker from responsibility for his abuse of that right.”); Kennedy v. Sheriff of E. Baton Rouge, 935 So.2d 669, 680 (La.2006) (“Courts in other states with similar clauses in their constitutions have interpreted the proviso against abuse as evidencing an express concern for injury to reputation that justifies adoption of a negligence standard for private plaintiffs in defamation actions. We agree with this line of cases and will not ignore the express concern for injury to reputation found in the Louisiana Constitution.” (Citations omitted.)); Martin v. Griffin Television, Inc., 549 P.2d 85, 91 (Okla.1976) (“Expressly in its constitution, Oklahoma has weighted the right with the responsibility for an abuse of that right. That same responsibility is not expressly found in the federal constitution.”) Kentucky has a similar abuse clause to ours and, like us, has adhered to the common law defamation presumptions. McCall v. Courier-Journal & Louisville Times Co., 623 S.W.2d 882, 886 (Ky.1981) (adopting a negligence standard for claims against media defendants in light of United States Supreme Court precedent and the state’s abuse clause while reaffirming “the basic common law and statutory rules of libel and slander as expressed and interpreted by this court in the past”); Columbia Sussex Corp. v. Hay, 627 S.W.2d 270, 273 (Ky.Ct.App.1981) (noting that defamation imposes strict liability on a per se theory and that “the defamatory utterance is presumptive evidence of the injury to reputation and of the ill will otherwise necessary to support a punitive award”); see also Kanaga v. Gannett Co., 687 A.2d 173, 181—82 (Del.1996) (noting Delaware’s constitutional abuse clause, similar to Iowa’s, and going on to recognize presumed damages for private plaintiff/nonmedia defendant cases).
A Utah Supreme Court decision presents some historical context for this type of constitutional language. See Am. Bush v. City of South Salt Lake, 140 P.3d 1235, 1244-53 (Utah 2006) (holding that the Utah Constitution does not protect nude dancing in light of the abuse clause). In analyzing the speech component of the *453Utah Constitution, which is substantially similar to Iowa’s, that court found that “the phrase ‘responsible for the abuse’ was intended to preserve liability for defamation” among other things. Id. at 1241, 1244 (alterations omitted) (citation and internal quotation marks omitted). The court noted that, following the revolutionary period in the United States, states consciously replaced broader speech guarantees based on Cato’s Letters with a more restrictive model inspired by Blackstone. Id. at 1248 (“[W]hen the Utah framers chose to include [an abuse clause], they chose a phrase with a long history of preserving the power of the state to regulate speech under certain historical exceptions.”). The states’ adoption of the Black-stonian model, emphasizing accountability for “improper, mischievous, or illegal” speech, was a response to the unyielding protection offered by Cato’s Letters. Id. (citation and internal quotation marks omitted); see also Miami Herald Publ’g Co. v. Ane, 458 So.2d 239, 241 (Fla.1984) (stating that “Florida’s concern for individual reputation is reflected in” the wording of Florida’s free speech protection); Bentley v. Bunton, 94 S.W.3d 561, 578 (Tex.2002) (examining Texas’s nearly identical abuse clause, and noting “[i]f anything, in the context of defamation, the First Amendment affords more protection”).
Although a majority of states have some form of abuse clause, there is not always consensus on the meaning and scope of that language. See 1 Jennifer Friesen, State Constitutional Law: Litigating Individual Rights, Claims, and Defenses § 5.02[3][e], at 5-10 (4th ed.2006); see also Lawson v. Helmer, 77 P.3d 724, 728 (Alaska 2003) (holding an absolute privilege could apply to in-court testimony and stating that “[i]n providing that citizens are responsible for abusing their right to free speech, the Alaska Constitution did not create an absolute right to sue others for defamation”); Yetman v. English, 168 Ariz. 71, 811 P.2d 323, 334 (1991) (noting Arizona’s abuse clause and stating that “whatever its scope of application in other areas, [Arizona’s free speech guarantee] provides no greater privilege for otherwise defamatory statements than the first amendment of the United States Constitution”); Degrassi v. Cook, 29 Cal.4th 333, 127 Cal.Rptr.2d 508, 58 P.3d 360, 364 (2002) (noting that the abuse clause “implicitly contemplated the continued existence of a long-established common law action for defamation”); Price v. State, 622 N.E.2d 954, 964 (Ind.1993) (“When the expressions of one person cause harm to another in a way consistent with common law tort, an abuse under [the state speech protection] has occurred.”); Bradburn v. N. Cent. Reg’l Library Dist., 168 Wash.2d 789, 231 P.3d 166, 172 (2010) (under the state abuse clause “no greater protection [than under the First Amendment] is afforded to obscenity, speech in nonpublic forums, commercial speech, and false or defamatory statements”). Yet in surveying this array of cases from around the country, we do not find much if any support for ASI and Scott’s contention that the abuse clause renders libel per se unconstitutional.
In sum, we are unable to conclude that the Iowa Constitution — a document that acknowledges criminal libel and liability for abuse of speech — provides defendants in defamation cases with more protection than the United States Constitution.
3. Policy considerations. Regardless of what the Federal or Iowa Constitution may require, ASI and Scott also urge us to abandon libel per se on policy grounds. They contend that the doctrine is outmoded and that a distinction between media and nonmedia defendants is no longer workable. We disagree on both scores.
*454We believe that libel per se remains a useful rule in an area where it is often difficult for a plaintiff to prove actual damages:
The rationale of the common-law rules has been the experience and judgment of history that proof of actual damage will be impossible in a great many cases where, from the character of the defamatory words and the circumstances of publication, it is all but certain that serious harm has resulted in fact.... As a result, courts for centuries have allowed juries to presume that some damage occurred from many defamatory utterances and publications.
Dun & Bradstreet, 472 U.S. at 760-61, 105 S.Ct. at 2946, 86 L.Ed.2d at 603 (citation and internal quotation marks omitted).
The harm resulting from an injury to reputation is difficult to demonstrate both because it may involve subtle differences in the conduct of the recipients toward the plaintiff and because the recipients, the only witnesses able to establish the necessary causal connection, may be reluctant to testify that the publication affected their relationships with the plaintiff. Thus some presumptions are necessary if the plaintiff is to be adequately compensated.
Note, Defamation, 69 Harv. L.Rev. 875, 891-92 (1956). These observations, we believe, remain valid today.
ASI and Scott argue that the Internet is “a great equalizer” and has rendered libel per se obsolete because the targets of defamation can respond quickly at minimal cost. We are not persuaded, however, that the Internet’s ability to restore reputations matches its ability to destroy them. As the New Jersey Supreme Court recently explained:
In today’s world, one’s good name can too easily be harmed through publication of false and defaming statements on the Internet. Indeed, for a private person defamed through the modern means of the Internet, proof of compensatory damages respecting loss of reputation can be difficult if not well-nigh insurmountable. We question why New Jersey’s longstanding common law tradition of presumed damages — for defamation claims by private citizens on matters that do not involve public concern— should be altered now to force an average citizen to ferret out proof of loss of reputation from any of the world-wide potential viewers of the defamatory Internet transmission about that otherwise private person. We are not persuaded that the common law of this state need change to require such victims to demonstrate compensatory losses in order to proceed with a cause of action.
In sum, private persons face the real risk of harm through the modern ease of defamatory publications now possible through use of the Internet. Presumed damages vindicate the dignitary and peace-of-mind interest in one’s reputation that may be impaired through the misuse of the Internet. Permitting rep-utational damages to be presumed in a defamation action arising in that setting serves a legitimate interest, one that ought not be jettisoned from our common law.
W.J.A. v. D.A., 210 N.J. 229, 48 A.3d 1148, 1159-60 (2012).7
ASI and Scott also argue that libel per se recently has become subject to so many *455exceptions that it is not worth preserving what remains. In making this argument, they point to Barreca. We have trouble following their position. Barreca did not carve out a new exception to defamation per se. Instead, it reaffirmed the doctrine of slander per se. Barreca, 683 N.W.2d at 116. It applied the longstanding defense of qualified privilege to statements made by an alderman at a city council meeting. Id. at 119. It attempted to eliminate some “confusion” in the caselaw about when the privilege exists and what must be shown to overcome it. Id. at 117-23. Barreca provided helpful clarification (we believe), but it did not significantly change the law. Defendants do not contend that a qualified privilege applies in this case.
This case illustrates why retaining libel per se for private plaintiff/private concern/nonmedia defendants may be appropriate. In our present-day world, accusations can be spread quickly and inexpensively, through self-publishing of a book or otherwise. A generation or two ago, it is entirely plausible that if Scott had decided to write a memoir about his life, it would have stayed by his typewriter and never been copied or distributed. Now, however, for a relatively modest price, it is possible to print 250 copies of a professional-looking book alleging that one’s ex-wife is a victim of child abuse from her father. We think libel per se plays a useful role in helping to keep our social interactions from becoming ever more coarse and personally destructive.
We are not persuaded that debate and discussion are insufficiently robust in Iowa, or that libel jury verdicts and the costs of defending libel actions are a drag on free speech in this state, or that Iowa has less vibrant discourse when compared with other states that have abolished libel per se.
Iowa is not the only state to continue to apply common law per se presumptions in private plaintiff/private concern cases involving (at least) nonmedia defendants. See Delta Health Grp., Inc. v. Stafford, 887 So.2d 887, 896 (Ala.2004) (“Damage is implied by law when spoken words are found to be slander per se.”); MacDonald v. Riggs, 166 P.3d 12, 15 (Alaska 2007) (noting that libel and slander per se do not require proof of special damages); Denver Publ’g Co. v. Bueno, 54 P.3d 893, 900 (Colo.2002) (“[I]f the plaintiff is a private person, and the claim is for libel per se, the plaintiff need not prove actual damages.”); Gaudio v. Griffin Health Servs. Corp., 249 Conn. 523, 733 A.2d 197, 215 (1999) (finding that reputational injury could be conclusively presumed in a defamation action by an employee against an employer for claims made in a termination letter); Twite v. Corbitt, 224 Ill.2d 490, 310 Ill.Dec. 303, 866 N.E.2d 114, 121 (2006) (“A statement is defamatory per se if its defamatory character is obvious and apparent on its face and injury to the plaintiffs reputation may be presumed.”); Baker v. Tremco Inc., 917 N.E.2d 650, 657 (Ind.2009) (stating that in a defamation per se action, no proof of injury is required); Stringer v. Wal-Mart Stores, Inc., 151 S.W.3d 781, 793-94 (Ky.2004) (indicating that Kentucky follows a traditional common law approach to defamation per se in private plaintiff/private concern cases and that damages and malice are presumed); Costello v. Hardy, 864 So.2d 129, 140 (La.2004) (stating that “[w]hen a plaintiff proves publication of words that are defamatory per se, the elements of falsity and malice (or fault) are presumed, but may be rebutted by the defendant” and “[t]he element of injury may also be presumed”); Morgan v. Kooistra, 941 A.2d 447, 455 (Me.2008) (indicating that defamation action requires “either actionability of the statement irrespective of special harm *456or the existence of special harm caused by the publication”); Mitan v. Campbell, 474 Mich. 21, 706 N.W.2d 420, 421 (2005) (indicating that defamation per se renders a statement actionable “irrespective of special harm”); State v. Crawley, 819 N.W.2d 94, 104 (Minn.2012) (stating that Minnesota recognizes defamation per se which is “actionable without any proof of actual damages” (citation and internal quotation marks omitted)); Speed v. Scott, 787 So.2d 626, 632 (Miss.2001) (explaining that no proof of special harm is required for slander per se); Blue Ridge Homes, Inc. v. Thein, 345 Mont. 125, 191 P.3d 374, 382 (2008) (“Defamation per se requires no proof of special damages.”); McCune v. Neitzel, 235 Neb. 754, 457 N.W.2d 803, 810 (1990) (“In a suit for slander per se, no proof of any actual harm to reputation or any other damage is required for the recovery of either nominal or substantial damages.”); Bongiovi v. Sullivan, 122 Nev. 556, 138 P.3d 433, 448 (2006) (finding a doctor’s statement that another doctor had poor surgical skills was slander per se and the plaintiff was entitled to presumed damages); Lassonde v. Stanton, 157 N.H. 582, 956 A.2d 332, 342 (2008) (upholding a damage award to a home contractor in a defamation per se case despite the absence of proof of damages); Geraci v. Probst, 15 N.Y.3d 336, 912 N.Y.S.2d 484, 938 N.E.2d 917, 922 (2010) (stating that damages will be presumed “for statements that charge a person with committing a serious crime or that would tend to cause injury to a person’s profession or business”); Ellis v. N. Star Co., 326 N.C. 219, 388 S.E.2d 127, 129 (1990) (noting that the court has previously held that in libel per se actions, damages may be presumed without a finding of malice); Brown v. Gatti, 341 Or. 452, 145 P.3d 130, 133 (2006) (explaining that certain defamation is actionable without proof of specific harm); Nassa v. Hook-SupeRx, Inc., 790 A.2d 368, 374 (R.I.2002) (“For slander per se, a plaintiff can establish liability without a showing of special or pecuniary damages because those damages are presumed.”); Fountain v. First Reliance Bank, 398 S.C. 434, 730 S.E.2d 305, 309 (2012) (indicating that when a statement is defamatory per se, the defendant “is presumed to have acted with common law malice and the plaintiff is presumed to have suffered general damages” (citation and internal quotation marks omitted)); Salinas v. Salinas, 365 S.W.3d 318, 320 (Tex.2012) (“Our law presumes that statements that are defamatory per se injure the victim’s reputation and entitle him to recover general damages, including damages for loss of reputation and mental anguish.” (Citation and internal quotation marks omitted.)); Larson v. SYSCO Corp., 767 P.2d 557, 560 (Utah 1989) (indicating that with defamation per se, malice and damages are presumed); Askew v. Collins, 283 Va. 482, 722 S.E.2d 249, 251 (2012) (“[T]he jury needed no proof of damages suffered by Collins on which to predicate its compensatory award based upon the per se defamation.... ”); In re Judicial Disciplinary Proceedings Against Gableman, 325 Wis.2d 579, 784 N.W.2d 605, 624 (2010) (“A plaintiff in a traditional defamation action, unless proceeding on a theory of defamation per se, proves damages or a harm to reputation.”); Hoblyn v. Johnson, 55 P.3d 1219, 1233 (Wyo.2002) (“Defamation per se means a statement which is defamatory on its face and, therefore, actionable without proof of special damages.” (Citation and internal quotation marks omitted.)).8
*457Turning to the media/nonmedia distinction, it is true that a number of cases and commentators have criticized it, primarily from a constitutional rather than a common law standpoint. But the criticism is not new. See, e.g., Restatement (Second) of Torts § 580B cmt. e, at 225-26 (1977). And one premise of the criticism may no longer be as valid as it used to be. According to the drafters of the Second Restatement:
It would seem strange to hold that the press, composed of professionals and causing much greater damage because of the wider distribution of the communication, can constitutionally be held liable only for negligence, but that a private person, engaged in a casual private conversation with a single person, can be held liable at his peril if the statement turns out to be false, without any regard to his lack of fault.
Id. Thus, in 1977, the drafters of the Restatement believed the greater power of the institutional media to inflict harm counseled against giving it more legal protection.
In recent years, however, the Internet and social media have evened the playing field somewhat, by giving individuals with access to a computer a ready platform for spreading falsehoods or engaging in cyber-bullying. Yet unlike the media, these individuals may have fewer incentives to self-police the truth of what they are saying. For example, they may speak anonymously or pseudonymously. Also, because they are not in the communications business, they may care less about their reputation for veracity. In short, as compared to a generation ago, nonmedia defendants may have a greater capacity for harm without corresponding reasons to be accurate in what they are saying. This is a justification for retaining our media/nonmedia distinction.
Also, Congress has recognized a distinction with some parallels to the media/non-media distinction in the Communications Decency Act of 1996. See 47 U.S.C. § 230 (2006). That act insulates a provider of an interactive computer service from defamation liability for “information provided by another information content provider.” Id. § 230(c)(1); see also Carafano v. Metrosplash.com, Inc., 339 F.3d 1119, 1122 (9th Cir.2003) (“Congress granted most Internet services immunity from liability for publishing false or defamatory material so long as the information was provided by *458another party. As a result, Internet publishers are treated differently from corresponding publishers in print, television and radio.”). Hence, Congress concluded that one could legitimately distinguish among potential defamation defendants, by eliminating all liability for the service provider that merely passes along allegedly libelous material while allowing state law to continue to impose liability on the originator of that material. Of course, a media defendant is not necessarily the same as an interactive computer service provider, and vice versa,9 but the point remains that our national elected representatives as a matter of public policy decided that one who initially makes a false statement about another can be treated differently from an entity that merely enables that statement to be publicized.10
We have recognized libel per se continuously since the nineteenth century. See Morse v. Times-Republican Printing Co., 124 Iowa 707, 718, 100 N.W. 867, 871 (1904) (noting “the falsity of the defamatory matter, malice in its publication, and injury to the plaintiff are all presumed”); Scholl v. Bradstreet Co., 85 Iowa 551, 554, 52 N.W. 500, 501 (Iowa 1892) (“As the publication by the defendant was not actionable per se, it was incumbent on the plaintiff to prove that there were special damages, and that the publication was made in malice.”); Call v. Larabee, 60 Iowa 212, 215, 14 N.W. 237, 288 (Iowa 1882) (noting that where a libel is actionable per se, “[t]he law presumes that damages do result from the libel”). Our current Geriz-inspired framework, which distinguishes media and nonmedia defendants and reserves libel per se for private plaintiffs and nonmedia defendants, has endured since 1984. See Vinson, 360 N.W.2d at 118. Even before then, our defamation law afforded more leeway to media defendants in certain circumstances. See Cherry v. Des Moines Leader, 114 Iowa 298, 304, 86 N.W. 323, 325 (1901) (stating that “the editor of a newspaper has the right, if not the duty, of publishing, for the information of the public, fair and reasonable comments, however severe in terms, upon anything which is made by its owner a subject of public exhibition, as upon any other matter of public interest; and such a publication falls within the class of privileged communications, for which no action will lie without proof of actual malice”), abrogated in pari on other grounds by Barreca, 683 N.W.2d at 119-21; see also Haas v. Evening Democrat Co., 252 Iowa 517, 531, 107 N.W.2d 444, 453 (1961) (holding that a newspaper for the most part was entitled to a qualified privilege in a pre-New York Times v. Sullivan case and noting that “[a] man who commences a newspaper war cannot subsequently come to the court as plaintiff to complain that he has had the worst of the fray” (citation and internal quotation marks omitted)). But see Morse, 124 Iowa at 724, 100 N.W. at 873 (“The publisher of a newspaper possesses no immunity from liability on account of a libelous publication, not belonging to any other citizen.” (Citation and internal quotation marks omitted.)).
*459These precedents should not be tossed aside lightly.
“[T]he decision to make the paradigm shift that is caused by overruling established common-law principles must be tempered by judicial restraint, with deference to the doctrine of stare decisis and its role in perpetuating stability under the rule of law.”
State v. Becker, 818 N.W.2d 135, 160 (Iowa 2012) (quoting Aizupitis v. State, 699 A.2d 1092, 1094 (Del.1997)).
E. Is ASI a Media Defendant? Since we are retaining our common law of libel per se and our common law distinction between media defendants and other defendants for libel law purposes, we next need to consider whether ASI is a media defendant. The district court found that ASI was not a media defendant, reasoning, “ASI is not the New York Times, or any other media entity. Rather, it is a business which contracts to publish documents for private authors.”
We recognize that our recent defamation eases involving media defendants have dealt with the news media. Caveman Adventures, 633 N.W.2d at 761-62 (finding a showing of actual malice required to recover punitive damages against a newspaper); Schlegel, 585 N.W.2d at 219, 224-26 (applying a media defendant damages standard to a case against a newspaper publisher and an editor-in-chief); Nickerson, 542 N.W.2d at 510-12 (finding a newspaper was a media defendant and, thus, “the distinction between libel and libel per se has become irrelevant”); Jones, 440 N.W.2d at 888 (finding that a television news company was a media defendant).
But we do not believe the concept of a media defendant encompasses only businesses that report news. Rather, our purpose was to track Gertz’s definition of “media defendant.” See Caveman Adventures, 633 N.W.2d at 761; Vinson, 360 N.W.2d at 117-18. Gertz involved a monthly magazine. 418 U.S. at 325, 94 S.Ct. at 3000, 41 L.Ed.2d at 797. The United States Supreme Court repeatedly indicated in Gertz that its principles would apply to “a publisher or broadcaster.” See id. at 340, 94 S.Ct. at 3007, 41 L.Ed.2d at 805-06 (“Our decisions recognize that a rule of strict liability that compels a publisher or broadcaster to guarantee the accuracy of his factual assertions may lead to intolerable self-censorship.”); id. at 346, 94 S.Ct. at 3010, 41 L.Ed.2d at 809 (referring to “a publisher or broadcaster”); id. at 347, 94 S.Ct. at 3010, 41 L.Ed.2d at 809 (same); id. at 348, 94 S.Ct. at 3011, 41 L.Ed.2d at 810 (referring to “the press and broadcast media”).
Other jurisdictions have held that book publishers are media defendants. See Geiger v. Dell Publ’g Co., 719 F.2d 515, 516 (1st Cir.1983) (finding that a publisher of an autobiographical book is a media defendant for defamation purposes); Shaari v. Harvard Student Agencies, Inc., 427 Mass. 129, 691 N.E.2d 925, 928 (1998) (finding that the publisher of a travel guidebook was a media defendant); Main v. Royall, 348 S.W.3d 381, 387 (Tex.App.2011) (holding that a book publisher was a “member of the electronic or print media” for purposes of Texas law). The First Circuit in Geiger observed that book publishers should trigger the same constitutional protections as news outlets.
Although it is true that book publishers are not often under the sort of time pressure that requires them to commit a story to print within the space of a few hours, we note that they operate under economic constraints that prevent their conducting the kind of routine check appellant wishes us to impose on them. A non-fiction work often details events that are long past and describes people who are unavailable to verify the au*460thor’s statements. To require a book publisher to check, as a matter of course, every potentially defamatory reference might raise the price of nonfiction works beyond the resources of the average man. This result would, we think, produce just such a chilling effect on the free flow of ideas as First Amendment jurisprudence has sought to avoid.
Geiger, 719 F.2d at 518.
We believe these publishers are part of the “press” separately recognized by the First Amendment of the United States Constitution and article I, section 7 of the Iowa Constitution. See U.S. Const. amend. I (“Congress shall make no law ... abridging the freedom of speech, or of the press....”); Iowa Const. art. I, § 7 (“No law shall be passed to restrain or abridge the liberty of speech, or of the press.”). The press play a vital role in our country by regularly circulating ideas, whether in book, magazine, or newspaper form. Thus, to hold the press or one of its agents or employees legally liable for a statement, our precedents require something more than that the statement be libel per se.
ASI by its own admission is not a “traditional publisher.” But it provided several publishing services. It designed and physically produced the book. It distributed the book. Although ASI did not promote or line-edit the book, it did run a manuscript scrub software on it. The software searched for, among other things, passages that could be potentially obscene or defamatory. After the program was run on Scott’s book, an employee of ASI discussed a problem area with him. Additionally, ASI acknowledged that it has dedined to publish books in the past, because they raise libel, copyright, trademark, or even morality concerns. A person or entity like ASI whose regular practice is to (1) receive written materials prepared by a number of different third parties and (2) make finished products from the materials that are designed to be more suitable and accessible for the public to read should be considered a publisher and a media defendant for purposes of our case law.11 See Parisi v. Sinclair, 774 F.Supp.2d 310, 320 n. 6 (D.D.C.2011) (describing ASI’s role in another case as that of a publisher).
Furthermore, the plaintiffs concede that if ASI were just a contract printer, it could not be found liable without proof of negligence. See Maynard v. Port Publ’ns, Inc., 98 Wis.2d 555, 297 N.W.2d 500, 507 (1980); Restatement (Second) of Torts § 581, at 231 (providing that one who merely delivers or transmits defamatory matter published by a third person is subject to liability only if he knows or has reason to know of its defamatory character). For example, in Sandler v. Calcagni, the plaintiff brought a libel action against “Book-Surge,” a self-publishing company that allows authors to upload manuscripts on its website and transform them into bound books. 565 F.Supp.2d 184, 187 (D.Me.2008). Unlike ASI, BookSurge “does not review submissions for content.” Id. BookSurge’s employees “did not read or review the manuscript submitted to them.” Id. at 190. Although BookSurge did not market the book, it was available for purchase through Amazon.com. Id. at 190-91. The district court analogized BookSurge to a contract printer and granted summary judgment in its favor, finding no evidence of negligence.12 Id. at 196.
*461And the plaintiffs concede that if ASI were a traditional publisher, liability could not be imposed without fault. Plaintiffs’ argument seems to be that because ASI falls into some kind of “no man’s land” in between contract printer and traditional publisher, it can be subject to strict liability. This strikes us as incongruous. Companies would have an incentive either to do no independent review of what they are publishing or a great deal of review. However, a company like ASI that tried to meet market demand by providing a more limited package of services would lose “publisher” status. Since what really matters is whether the entity is regularly engaged in the dissemination of an author’s ideas through outlets that are not otherwise readily available to that author, it makes no sense to draw lines of this kind.13
' It is true that Scott paid ASI to publish his book, rather than the other way around. But this fact alone does not change the analysis. Both our precedents and the United States Supreme Court’s have accorded the same protection to media defendants when they publish advertisements as when they publish content they have paid for. See Caveman Adventures, 633 N.W.2d at 761-62; see also New York Times, 376 U.S. at 256, 264, 84 S.Ct. at 713, 717, 11 L.Ed.2d at 692, 697.
We believe that following our established defamation law and recognizing ASI as a media defendant will afford adequate protection to individuals who have been victimized by the false statements of others. The plaintiffs can still pursue a libel per se claim against Scott, because he is not a media defendant.14 In this regard, we take note of a recent decision that bears considerable resemblance to the present case. See Lassiter v. Lassiter, 456 F.Supp.2d 876 (E.D.Ky.2006), aff'd, 280 Fed.Appx. 503 (6th Cir.2008). In Lassiter, the defendant published a book entitled I Have a Testimony following her divorce from her husband. As explained by the district court:
The book is primarily of an inspirational and religious nature. Ms. Lassi-ter’s main theme in the book is how her faith and the power of prayer have seen her through many trying times in her life, including certain phases of her marriage and her divorce.
She alleges ongoing mental cruelty and abuse by her husband throughout the marriage. She also states in the book that he physically assaulted her on two occasions. Further, she charges in the book that her husband committed adultery during the marriage.
Although defendant did not mention plaintiff by name in the book, everybody who knew the couple knew to whom she was referring when she referred to her husband.
Id. at 878. The district court applied Kentucky’s law of libel per se, recognizing that this “is an action by one private person against another private person about a matter that is not of public interest.” Id. at 880. Significantly, it also rejected the *462defendant’s claim that she was a “media defendant” even though she “self-published her book with her personal computer.” Id. at 878 n. 1, 880. Ultimately, though, the district court found for the defendant because she demonstrated that the defamatory statements were either true or protected opinions as to which the underlying facts had been disclosed. Id. at 879-80, 882. A key difference between Lassiter and the present case, of course, is that Ms. Lassiter published the book on her own; but a key similarity between both cases is that the underlying author of the allegedly defamatory material remains subject to a potential libel per se cause of action.
F. Do the Challenged Statements Involve Matters of Public Concern? ASI and Scott also argue that libel per se cannot be applied to them because the allegedly defamatory statements in the book involve matters of public concern. See Dun & Bradstreet, 472 U.S. at 759-63, 105 S.Ct. at 2945-47, 86 L.Ed.2d at 602-05; Barreca, 683 N.W.2d at 120 n. 6. This argument need not be reached as to ASI because we have already concluded that it is a media defendant. As to Scott, we disagree.
Public concern protection serves the constitutional goal of “assuring] unfettered interchange of ideas for the bringing about of political and social changes desired by the people.” Dun & Bradstreet, 472 U.S. at 759, 105 S.Ct. at 2945, 86 L.Ed.2d at 602 (citation and internal quotation marks omitted). On the other hand, “purely private disputes such as a lawsuit in which the impact is limited primarily to the parties involved, even though perhaps of interest to the public, are insufficient to create a matter of public concern.” Nickerson, 542 N.W.2d at 511.
Scott argues that Mind, Body and Soul relates to matters of public concern, namely, religious beliefs, mental health, and unprosecuted child abuse. We believe this approach would broaden the “public concern” category so it covers virtually anything. No man is an island, and everyone’s life is potentially of interest to everyone else. But the events described in the book would not reasonably be expected to have an impact beyond the parties involved. See Jones, 440 N.W.2d at 900. They take on broader significance only to the extent Scott has written about them and urged us to learn lessons from them. Accordingly, we do not believe there is any constitutional or common law bar to applying libel per se to Scott. See Lassiter, 456 F.Supp.2d at 880 (finding that a woman’s self-published “religious and inspirational” book accusing her ex-husband of abuse and adultery was not a matter of public concern); W.J.A., 43 A.3d at 1157 (rejecting the defendant’s claim that an allegation of child molestation was a matter of public concern).
G. Should ASI Have Been Granted Summary Judgment on the Plaintiffs’ Libel Claims? We turn now to whether ASI should have been granted summary judgment as a media defendant. As we have discussed, “presumed damages” are impermissible against a media defendant. Schlegel, 585 N.W.2d at 222. “Hurt feelings alone cannot serve as the basis of a defamation action.” Nickerson, 542 N.W.2d at 513. There must be proof of “reputational harm.” Schlegel, 585 N.W.2d at 224.
ASI moved for summary judgment below based on the absence of injury to reputation. It now reurges on appeal that it was entitled to summary judgment on this ground.
We believe our decision in Schlegel is on point. In that case, a newspaper incorrectly reported that a lawyer had declared bankruptcy. Id. at 220. The record was *463rife with evidence of hurt feelings and depression, but did not demonstrate that anyone thought less of the attorney. Id. at 225. “The Schlegels presented a number of witnesses, most of whom were friends, who saw the false report. None testified that Richard had any particular reputation before the false report or that they thought ill of him because of it.” Id. We held the defendants should have been granted judgment n.o.v. Id. at 226.
Plaintiffs’ case here suffers from the same gap in proof. While the summary judgment record contains evidence of the good reputations of both Beth and Gail before the publication of the book, it is devoid of evidence that anyone changed his or her opinion of the two after reading the book. The affidavits of friends revealed either that they had not read the book, or that if they had read portions of it, they did not accept the allegations it contained about Beth and Gail. Beth’s work supervisor averred that Beth has suffered mental anguish and was less outgoing at work than before the book was published — i.e., the same kind of proof we found insufficient in Schlegel — but he did not assert that anyone at work thought less of her because of the statements in the book. Beth testified that she did not know who might think less of her because of the publication of the book. She speculated that some people of whom she is not aware might have read it, and expressed her belief that it is just as likely someone thinks less of her after reading the book as it is that someone else read it and does not. Gail testified that he suffered stress because of the publication of Scott’s book, mainly because he had to endure a deposition. He did not identify anyone who believed the allegations about him published in the book and consequently thought less of him.
Beth and Gail urge us to consider the testimony of Scott’s parents who testified that they thought less of Beth and Gail because of the allegations of sexual abuse and mental illness. However, the record is clear that Scott’s parents formed their opinions about the abuse and the mental illness long before Scott wrote or published his book. They testified that they came to believe the abuse had occurred and that Beth suffered from some sort of personality disorder while Scott and Beth were still married, at least six years before the book was published. Accordingly, the testimony of Scott’s parents is not evidence tending to prove the publication of the statements in the book caused Beth and Gail reputa-tional harm.
Plaintiffs also argue we can infer reputa-tional harm circumstantially based on the evidence that (1) Beth and Gail had good reputations before the book was published and (2) approximately thirty-four copies of the book were distributed by gift or sale. Plaintiffs cite Wilson v. IBP, Inc., 558 N.W.2d 132 (Iowa 1996), and Lara v. Thomas, 512 N.W.2d 777 (Iowa 1994), in support of this argument. Both of these cases, however, were slander per se cases. Wilson, 558 N.W.2d at 140; Lara, 512 N.W.2d at 780. The fighting issue in these cases was whether the damages were excessive. 558 N.W.2d at 140, 512 N.W.2d at 780. Neither case holds that reputational harm, when proof of such harm is required, can be inferred from mere distribution of a publication. Indeed, allowing such an inference would in effect turn libel per quod into “libel per se lite.”
For the foregoing reasons, we conclude that ASI should have been granted summary judgment on plaintiffs’ libel claims.
H. Should Scott Have Been Granted Summary Judgment on the Plaintiffs’ Libel Claims? If Mind, Body and Soul amounts to libel per se, Beth and *464Gail need not prove any falsity, fault or reputational injury (damages) to proceed against Scott. However, they must meet other requirements. “To establish a prima facie case in any defamation] action, a plaintiff must show the defendant (1) published a statement that was (2) defamatory (3) of and concerning the plaintiff.” Taggart v. Drake Univ., 549 N.W.2d 796, 802 (Iowa 1996).
Scott first maintains there is no evidence that the allegedly libelous material was published to a third party. “Publication is an essential element of defamation and simply means a communication of statements to one or more third persons.” Huegerich, 547 N.W.2d at 221 (Iowa 1996). Scott argues, “The Plaintiffs simply cannot provide evidence of one single person who read any of the statements claimed to be defamatory.” This is not accurate. Several persons, including Scott’s mother, testified to having read or at least skimmed the book.
Scott next argues that the district court erred in finding as a matter of law that statements from Mind, Body and Soul were so defamatory that they amounted to libel per se. “If a statement is clear and unambiguous, the issue of whether the statement is libelous per se is for the court.” Kiesau, 686 N.W.2d at 175. Thus, the court may find a statement is libel per se if it unambiguously tends “to provoke the plaintiff to wrath or expose him to public hatred, contempt, or ridicule.” Nickerson, 542 N.W.2d at 510. Accusations of indictable crimes of moral turpitude are libel per se. See Huegerich, 547 N.W.2d at 221 (accusing plaintiff of possessing illegal drugs is libel per se); Rees v. O’Malley, 461 N.W.2d 833, 835 (Iowa 1990) (accusing plaintiff of extortion is libel per se); Vinson, 360 N.W.2d at 115-16 (accusing plaintiff of falsifying time cards is libel per se). Likewise, an accusation of immorality or dishonesty is libel per se. See Kiesau, 686 N.W.2d at 178 (stating that substantial evidence supported a jury finding that a doctored image of plaintiff appearing topless was libel per se); Wilson, 558 N.W.2d at 139-40 (stating that an accusation of untruthfulness was sufficient evidence to support a jury finding of libel per se).
We agree with the district court that “stating a person has been molested by their father and suffers from bipolar disorder constitutes libel per se under Iowa law.” Obviously, this does not preclude Scott from raising other defenses (such as truth, which he has pled in his answer). It means only that certain statements in the book are of a character that our common law views as libelous per se.15
Additionally, Scott claims the identified statements cannot be considered libel per se because they were not “of and concerning” the plaintiffs. As Scott points out, Beth and Gail are not named in those statements. For example, in one passage we have previously noted, Scott wrote, “The two women we spoke of earlier, they were both molested by their fathers, or at least that is what they told me.” Scott argues that because one would have to resort to outside facts, that passage cannot be defamatory.
However, this element only requires that a third-party recipient be able to understand who is the intended subject. See Restatement (Second) of Torts § 564, cmt. a. Iowa Code section 659.1 provides, “In an action for slander or libel, it shall *465not be necessary to state any extrinsic facts for the purpose of showing the application to the plaintiff of any defamatory matter out of which the cause of action arose.... ” Iowa Code § 659.1 (2009). The necessary implication of this statutory language is that a libel action can be pursued even when “extrinsic facts” are required. See Boardman & Cartwright v. Gazette Co., 225 Iowa 533, 538, 281 N.W. 118, 120 (1938) (“Of course, it is not necessary to constitute a libel that the article name the person libeled, but it must by inference or innuendo at least refer in an intelligent way to the person libeled.”); see also Ruzieka v. Conde Nast Publ’ns, Inc., 999 F.2d 1319, 1322 n. 6 (8th Cir.1993) (“The plaintiff need not be cited by name for the defamation to be ‘of and concerning the plaintiff.’ ” (Citation omitted.)).
Here the passage in question, referring to “two women we spoke of earlier,” appears on page 20 of the book. In the preceding nineteen pages, only two women are discussed — Scott’s ex-wife and a woman who became pregnant and claimed Scott was the father. As the district court found, it does not take speculation or guesswork to put two and two together. Other statements that are the subject of the lawsuit clearly refer to Scott’s “ex” or “ex-wife.” Accordingly, the district court correctly denied Scott’s motion for summary judgment, because Beth and Gail have shown the existence of a fact issue as to whether the challenged statements were “of and concerning” them.
I. Should ASI Have Been Granted Summary Judgment on the Plaintiffs’ False Light Invasion of Privacy Claim? ASI contends the district court should have granted its motion for summary judgment on plaintiffs’ claim of false light invasion of privacy. This claim arises in the following circumstances:
One who gives publicity to a matter concerning another that places the other before the public in a false light is subject to liability to the other for invasion of his privacy, if (1) the false light in which the other was placed would be highly offensive to a reasonable person, and (2) the actor had knowledge of or acted in reckless disregard as to the falsity of the publicized matter and the false light in which the other would be placed.
Kiesau, 686 N.W.2d at 179 (quoting Winegard v. Larsen, 260 N.W.2d 816, 823 (Iowa 1977)).
In the proceedings below, ASI argued, among other things, that the required evidence of malice was lacking. The district court denied ASI’s motion on this point without extensive analysis, simply observing that a fact issue was presented whether ASI “had knowledge of or acted in reckless disregard of the falsity of the publicized matter.” We respectfully view the matter otherwise.
As Beth and Gail note, ASI adopted a computerized process for review of certain buzz words, which it used on the text, but then did no further follow-up when the program retrieved the aforementioned “two women we mentioned earlier” passage. We do not believe this evidence, viewed most favorably to the plaintiffs, constitutes legal malice. There is no indication in the record that ASI doubted the veracity of the book or had a specific reason to do so. See Harte-Hanks Commc’ns, Inc. v. Connaughton, 491 U.S. 657, 688, 109 S.Ct. 2678, 2696, 105 L.Ed.2d 562, 589 (1989) (“There must be sufficient evidence to permit the conclusion that the defendant in fact entertained serious doubts as to the truth of his publication.” (Citation and internal quotation marks omitted.)). Although Scott’s book vented at times about his ex-wife, it did not do so to *466a degree or in ways that would have put ASI on notice he was making things up. Much of the book consists of Scott’s religious reflections. Accordingly, we conclude the district court should have granted summary judgment to ASI on plaintiffs’ false light invasion of privacy claim.
J. Should Scott Weier Have Been Granted Summary Judgment on the Plaintiffs’ False Light Invasion of Privacy Claim? Scott also contends the district court should have granted summary judgment in his favor on the false light claims. He first asserts that Beth and Gail have failed to establish the “publicity” element of the claim.
“Publicity” [for an invasion of privacy claim] means that the matter is made public, by communicating it to the public at large, or to so many persons that the matter must be regarded as substantially certain to become one of public knowledge. The difference is not one of the means of communication, which may be oral, written or by any other means. It is one of a communication that reaches, or is sure to reach, the public.
Thus it is not an invasion of the right of privacy ... to communicate a fact concerning the plaintiffs private life to a single person or even to a small group of persons. On the other hand, any publication in a newspaper or a magazine, even of small circulation, or in a handbill distributed to a large number of persons, or any broadcast over the radio, or statement made in an address to a large audience, is sufficient to give publicity within the meaning of the term.... The distinction, in other words, is one between private and public communication.
Restatement (Second) of Torts § 652D cmt. a, at 384-85.
We think the record raises a fact question as to whether the book and the allegations contained therein were sufficiently publicized to preclude summary judgment on these claims. Approximately twenty to thirty copies of the book were distributed. Scott actually ordered 250 copies of the book and attempted to market the book for sale at local businesses. He participated in a television interview promoting the book. As noted by the district court, the book was “available for purchase on the world-wide web for a period of approximately two months.” The record indicates several people have read the book or portions of the book. The finder of fact will need to decide if this is enough publicity to sustain a false light invasion of privacy claim.
We also think Beth and Gail have engendered a fact question on the malice element. The allegations made in the book are based on interactions and conversations that allegedly occurred between Beth and Scott during their marriage. If what Scott says is untrue, there is certainly a fact issue as to whether he knew it was untrue or acted with reckless disregard of its truth or falsity. Accordingly, we affirm the district court’s denial of Scott’s motion for summary judgment on this claim.
K. Should Scott Weier Have Been Granted Summary Judgment on the Plaintiffs’ Intentional Infliction of Emotional Distress Claims? Scott asserts Beth and Gail have failed to establish a triable issue on the outrageous conduct element of their intentional infliction of emotional distress claims. See Barreca, 683 N.W.2d at 123 (listing “outrageous conduct by the defendant” as one element of the tort). He contends that because he did not refer to Beth or Gail by name in the critical parts of the book or in any publicity, his actions do not constitute the outrageous conduct necessary to support an intentional infliction of emotional distress claim. As Scott puts it, “[I]f Appel*467lants had wanted to maliciously or with reckless disregard bring about unwarranted publicity or intrude on Appellees, they would have certainly mentioned them or referred to them in the very limited ‘publicity’ the hook did get.” Evaluating these contentions alone, and assuming all factual disputes are resolved in Beth and Gail’s favor, we find a triable issue of fact as to whether Scott’s conduct was “outrageous.”
Accordingly, we affirm the district court’s denial of Scott’s motion for summary judgment on this claim.
IV. Conclusion.
For the reasons described above, we reverse the district court’s denial of ASI’s motion for summary judgment as to the plaintiffs’ libel and false light invasion of privacy claims and remand for entry of judgment in ASI’s favor. We affirm the district court’s denial of summary judgment for Scott on the libel, false light, and intentional infliction of emotional distress claims, and remand for trial on those counts.
AFFIRMED IN PART, REVERSED IN PART, AND REMANDED FOR FURTHER PROCEEDINGS.
All justices concur except WIGGINS, J., who concurs specially, and HECHT and APPEL, JJ., who concur in part and dissent in part.. Because one of the plaintiffs and one of the defendants share the same last name, we will refer to all individual parties by their first names.
.The portions of the book identified by Beth and Gail as defamatory included the following passages:
• The two women we spoke of earlier, they were both molested by their fathers, or at least that is what they told me. And both of them were bipolar or borderline personality disorder, which is a fairly normal result of this type of sin against a child.
• Yes there are many skeletons in the closet that could be brought to the forefront, ones that would greatly damage my ex and her family.
• I have been held in contempt for things that are ridiculous, all because my ex has hired good lawyers who have gotten up in court and flat out lied, and in the end I have had to pay for her lawyers a few times to tell all these lies about me.
• Basically she has taught them a sinful nature, instead of the kingdom of God.
• Now you have just one parent trying to instill morals and values into my daughters. And what is she instilling? Hatred, non-forgiveness, no Christian morals (they do not go to church).
. Beth and Gail originally named Amazon.com as a defendant but later dismissed it with prejudice.
. Scott also filed counterclaims against Beth and Gail, alleging they had committed slander and slander per se against him in statements they made to his daughters. The district *442court granted Beth and Gail's motion for summary judgment on those counterclaims, and Scott has not challenged their dismissal on appeal.
. In Vinson, we expressly distinguished Anderson v. Low Rent Housing Commission, which ASI cites to us. 360 N.W.2d at 118 (distinguishing Anderson, 304 N.W.2d 239). Anderson involved a suit brought by a terminated city employee who was "in the midst of a controversy, which received widespread coverage from the news media, involving city projects, officials, and fellow employees.” 304 N.W.2d at 242. We held that a malice standard applied to the fired employee’s defamation claims against all defendants, both media and nonmedia, stating that "we find no basis in the plain language of the first amendment that would justify according greater protection to the media than to private parties.” Id. at 247. However, in Vinson we made clear that the Anderson outcome was based on the fact that the plaintiff was a public official. 360 N.W.2d at 118.
. ASI correctly notes that our criminal libel statute was repealed over a generation ago. See 1976 Iowa Acts ch. 1245, ch. 4 § 526. Still, we have continued to recognize libel per se. See, e.g., Kiesau, 686 N.W.2d at 176; Spencer, 479 N.W.2d at 296.
. The New Jersey Supreme Court did hold that presumed damages should be limited to "nominal damages” and that "[t]o receive a compensatory award for reputational loss, a plaintiff will be required to prove actual harm, pecuniary or otherwise, to his reputation through the production of evidence.” W.J.A., 43 A.3d at 1160.
. As the parentheticals indicate, many of these jurisdictions follow the Restatement approach and only presume damages, not falsity or malice, in the case of defamation per se. See Restatement (Second) of Torts, § 558 at 155 (1977). Several jurisdictions have cho*457sen to follow a different approach by abolishing defamation per se altogether. See, e.g., United Ins. Co. of Am. v. Murphy, 331 Ark. 364, 961 S.W.2d 752, 756 (1998) (adopting a rule that all defamation plaintiffs must establish actual reputational harm); Gobin v. Globe Pub. Co., 232 Kan. 1, 649 P.2d 1239, 1242 (1982) ("Damages recoverable for defamation may no longer be presumed; they must be established by proof, no matter what the character of the libel.”); Metromedia, Inc. v. Hillman, 285 Md. 161, 400 A.2d 1117, 1123 (1979) (stating that the libel per se only relieves the plaintiff from having to show the publication is defamatory); Nazeri v. Mo. Valley Coll., 860 S.W.2d 303, 313 (1993) (abandoning the libel per se/per quod distinction and requiring all libel plaintiffs to establish actual reputational harm to recover in a case involving a nonmedia defendant); Smith v. Durden, 276 P.3d 943, 948 (N.M.2012) (acknowledging abolition of distinction between libel per quod and libel per se in New Mexico and noting that key to analysis is the status of the plaintiff and holding that all defamation plaintiffs must establish actual harm to reputation to recover without consideration of the media/nonmedia status of defendant); Memphis Publ’g Co. v. Nichols, 569 S.W.2d 412, 419 (Tenn.1978) ("[T]he Per se/per quod distinction is no longer a viable one. The plaintiff must plead and prove injury from the alleged defamatory words, whether their defamatory meaning be obvious or not.”). Nonetheless, it would not be accurate to say that the concept of defamation per se is an outlier. It appears most jurisdictions continue to recognize some form of it.
. A newspaper reporter, for example, would be a media defendant even though she or he provides original content. However, a newspaper that publishes an advertisement, a radio broadcaster that puts a guest on the air, or a book publisher that prints someone else’s book, is in a similar position to the internet service provider under the CDA. See Jones, 440 N.W.2d at 888-90 (characterizing broadcast reporters as media defendants).
. Likewise, our general assembly has enacted defamation laws that distinguish between certain members of the news media and other defendants. See, e.g., Iowa Code §§ 659.2, .3, .5 (2009).
. To be clear, this is our test of when a person or entity becomes a bona fide publisher and therefore a media defendant.
. Maine law requires a defamation plaintiff to prove fault in every case. Sandler, 565 F.Supp.2d at 193. However, the significant point is that the court analyzed the case un*461der Restatement section 581. See id. at 193— 94.
. ASI puts the matter as follows:
ASI consistently argued to the District Court that it should be treated as a contract printer, with no prospect of liability, or in the alternative, as a media defendant who under the undisputed facts was not liable because of the lack of fault, falsity and damages. The District Court instead incorrectly accepted the argument of Appellees that ASI should be strictly liable. As a matter of law, ASI was entitled to status as one or the other, and in either case then was entitled to summary judgment. Appel-lees seek to deny it status as either.
. Scott is not an employee or agent of ASI.
. Neither party has asked us to dissect all the challenged statements in the book and determine whether they are — or are not — libelous per se. For purposes of the present appeal, we hold only that at least some of them are.