The plaintiff, Terry T. Thomas, appeals, and the defendants, Telegraph Publishing Company (Telegraph), Terrence L. Williams, Joshua Trudell, Town of Hudson (Town), Michael Gosselin, Roland Anderson, Albert Droney, Gene Bousquet, and Edith Flynn, cross-appeal an order of the Trial Court {Groff, J.) granting summary judgment in favor of the defendants. We affirm in part, reverse in part, vacate in part and remand.
I. Background
This case was the subject of a previous opinion of this court. See Thomas v. Telegraph Publ’g Co., 151 N.H. 435, 436 (2004). In December 2002, the plaintiff filed a civil action alleging defamation against the defendants based upon the publication of an article in the Nashua Telegraph on December 22, 1999. Id. It was entitled, “Police Say Burglar’s Luck Has Run Out After 25 Years.” The lengthy article states that Thomas “is now being held in the Hillsborough County House of Corrections on $25,000 bail while facing charges of receiving stolen property in Hudson.” The article also contains statements about the plaintiff’s past criminal behavior and indicates that he “is suspected in more than 1,000 home burglaries in Massachusetts and New Hampshire since the mid-1970’s, according to police and court records.” The article, in its entirety, is in an appendix to this opinion, as are the fifty-eight statements that the plaintiff challenges as defamatory.
The article was written by Trudell, and contained quotes or statements attributed to defendants Gosselin, Anderson, Droney, Bousquet and Flynn. At the time the article was written, Gosselin was a detective in Hudson, Anderson was the deputy police chief in Weston, Massachusetts, Droney was a detective in Needham, Massachusetts, and Bousquet was a detective in Foxborough, Massachusetts. Flynn was a professor of criminal justice at Northeastern University. Williams was the publisher of the Telegraph. In the remainder of this opinion, we reference Trudell, Williams and the Telegraph as “the Telegraph defendants.” We reference Gosselin, Anderson, Bousquet, Droney and the Town as “the police defendants.”
*320Over the course of litigating this case, the parties filed a number of motions with the trial court. The plaintiff moved to amend his writ to name the police officers in their individual — rather than just official — capacities. The trial court denied the motion, and the plaintiff appeals that decision.
Each of the defendants also moved for summary judgment. Over the plaintiff’s objection, the trial court granted summary judgment for all of the defendants on all of the allegedly defamatory statements, ruling that the plaintiff is libel-proof. In addition, the trial court ruled that: (1) certain statements in the article are covered by the fair report privilege; (2) certain of the statements are protected as substantially true; (3) Flynn’s statement was an opinion, but the statements of the police defendants were not statements of opinion; (4) Flynn’s statements were “of and concerning” the plaintiff; (5) the plaintiff is not a limited purpose public figure and therefore did not need to demonstrate actual malice; and (6) the police defendants do not enjoy a qualified privilege for their statements. The plaintiff then appealed, and the defendants cross-appealed, placing each of these rulings in dispute.
II. Motion to Amend
Arguing that his motion to amend was “a direct response” to discovery issues between the parties and an attempt to cure a technical defect, the plaintiff contends that the trial court erred in denying it. The decision of the trial court to deny a motion to amend will not be overturned absent an unsustainable exercise of discretion. Thomas, 151 N.H. at 439. Generally, a court should allow amendments to pleadings to correct technical defects, but need only allow substantive amendments when necessary to prevent injustice. Id. A substantive amendment that introduces an entirely new cause of action, or calls for substantially different evidence, may be properly denied. Id.
In response to the plaintiff’s motion to amend, the trial court made the following ruling:
The plaintiff filed this action over two years ago____His motion seeks to do more than cure a technical defect. He essentially seeks to add a number of new parties by suing the defendants in their individual capacities, as well as, their official capacities. The plaintiff previously represented to this Court that he would not be seeking any further amendments to the original Writ____ Given the delay in bringing this motion and the surprise to the defendants, the Court finds and rules that the amendment is not necessary for the prevention of injustice.
*321We agree with the trial court’s reasoning. By seeking to name the defendants in their individual capacities, the plaintiff essentially sought to add parties who would be personally liable for damages in the event of a verdict unfavorable to them. Further, defendants named in their individual capacities in this type of case might need to maintain defenses that would differ from those adopted by defendants named in their official capacities. These two considerations, among others, support the trial court’s conclusion that the defendants would suffer surprise from the proposed amendment. Clinical Lab Prod’s, Inc. v. Martina, 121 N.H. 989, 991 (1981) (surprise to opposing side is grounds to deny motion to amend). Indeed, the plaintiff filed the motion at issue some two years after he initiated the suit, after having previously amended his writ, and after having assured the trial court that no further amendments would be sought. Under these circumstances, we cannot conclude that the trial court unsustainably exercised its discretion in denying the plaintiff’s motion to amend.
III. Summary Judgment: Legal Standard
In acting upon a motion for summary judgment, the trial court is required to construe the pleadings, discovery and affidavits in the light most favorable to the non-moving party to determine whether the proponent has established the absence of a dispute over any material fact and the right to judgment as a matter of law. Porter v. Coco, 154 N.H. 353, 356 (2006). An issue of fact is material if it affects the outcome of the litigation. Id. We review de novo a trial court’s grant of summary judgment. Tech-Built 153 v. Va. Surety Co., 153 N.H. 371, 373 (2006). If our review of the evidence does not reveal any genuine issue of material fact, and if the moving party is entitled to judgment as a matter of law, we will affirm the trial court’s decision. Id.
IV. Libel-Proof Plaintiff Doctrine
The trial court granted summary judgment as to all defendants, ruling that the plaintiff is libel-proof. The plaintiff appeals that ruling.
Typically, “[a] plaintiff proves defamation by showing that the defendant failed to exercise reasonable care in publishing a false and defamatory statement of fact about the plaintiff to a third party, assuming no valid privilege applies to the communication.” Pierson v. Hubbard, 147 N.H. 760, 763 (2002). If defamation is established and no privilege applies, the plaintiff may seek damages for harm to his or her reputation. See Thomson v. Cash, 119 N.H. 371, 376 (1979). However, in some jurisdictions a very narrow class of plaintiffs is prohibited from seeking libel damages by operation of a doctrine known as the libel-proof plaintiff doctrine. See *322generally Annotation, Defamation: Who Is “Libel-Proof”, 50 A.L.R.4TH 1257 (1986 & Supp. 2006). Whether to adopt the libel-proof plaintiff doctrine in this state is a question of first impression.
One of the earliest formulations of the libel-proof plaintiff doctrine was announced by the United States Court of Appeals for the Second Circuit in Cardillo v. Doubleday & Co., Inc., 518 F.2d 638 (2d Cir. 1975); see also Note, The Libel-Proof Plaintiff Doctrine, 98 HARV. L. Rev. 1909, 1909-10 (1985). There, the court held that a prison inmate, who brought a civil libel action against the publisher of a book that referenced his involvement in various criminal organizations and activities, was libel-proof. Cardillo, 518 F.2d at 639-40. The court explained that the inmate was libel-proof because he was “so unlikely by virtue of his life as a habitual criminal to be able to recover anything other than nominal damages as to warrant dismissal of the case____” Id. at 639.
A. Incremental Harm Doctrine
Since Cardillo, two versions of the libel-proof plaintiff doctrine have developed. See Note, supra at 1910. One version is the incremental harm doctrine, which
involves an examination of the challenged communication rather than a finding of a previously damaged reputation. The judge evaluates the defendant’s communication in its entirety and considers the effects of the challenged statements on the plaintiff’s reputation in the context of the full communication. If the challenged statement harms a plaintiff’s reputation far less than unchallenged statements in the same article or broadcast, the plaintiff may be held libel-proof. Finding that the challenged statements could cause no cognizable damage in addition to that presumed to attend the unchallenged part of the communication, the court dismisses the entire libel action.
Id. at 1912-13.
In performing the above-described analysis, courts must ensure that the challenged statements are also actionable, since only actionable, challenged statements may be considered in assessing the harm done to a plaintiff’s reputation. Kite, Incremental Identities: Libel-Proof Plaintiffs, Substantial Truth, and the Future of the Incremental Harm Doctrine, 73 N.Y.U. L. Rev. 529, 548-63 (1998). Thus, if a challenged statement is nonactionable, it is treated like an unchallenged statement and its effects cannot be considered. Challenged statements “may be nonactionable for any number of reasons — constitutional defenses, substantial truth *323defenses, privileges allowed by state statute ....” Id. at 542-43. Although some courts have adopted the incremental harm doctrine, others have criticized and rejected it, and the United States Supreme Court has held that its application is not compelled by the First Amendment to the Federal Constitution. See id. at 548-63.
The Telegraph defendants contend that the incremental harm doctrine applies because “the accurate reporting of the plaintiff’s criminal record, most of which he has acknowledged in discovery, would have been more than sufficient to ‘demolish his reputation.’ Nothing published by The Telegraph in the Article did, or could have, done any incremental harm to this reputation above and beyond that which such an accurate report would have caused.” The police defendants also assert the applicability of the incremental harm doctrine. They appear to argue that it applies because the plaintiff admitted involvement in crime. We are not persuaded that the incremental harm doctrine applies in light of the record presented to us.
We acknowledge that the plaintiff admitted to several arrests and convictions during discovery in this case. Other arrests and convictions were deemed admitted by an order of the trial court, which the plaintiff does not here challenge. However, the plaintiff’s status as a convicted criminal and his admission to various criminal activities, alone, are not dispositive under the incremental harm doctrine. As stated above, the incremental harm doctrine focuses upon the communication at issue and the extent to which the challenged and actionable portions of its contents create harm above that caused by the portions that are unchallenged or nonactionable. The doctrine does not operate based upon a finding of a previously damaged reputation. See Note, swpra at 1912-13.
In addition, clearly some of the plaintiff’s arguments or challenges are undercut by the admitted facts; however, the article’s potential harm to the plaintiff’s reputation derives not only from admitted facts but also from other statements which, at this time, have not been either admitted as true or deemed nonactionable. Accordingly, this is not a case in which the plaintiff challenges only a small number of statements in an article that is, on the whole, otherwise unchallenged or nonactionable and largely injurious to his reputation. See Ferreri v. Plain Dealer Publishing Co., 756 N.E.2d 712, 723 (Ohio Ct. App. 2001). Rather, on the record presented to us in connection with this appeal, this is a case in which the plaintiff’s writ challenges fifty-eight out of ninety statements from the article, only some of which are nonactionable by operation of the fair report privilege, the substantial truth defense, and the opinion defense. None of the statements has yet been deemed qualifiedly privileged, a designation that would also *324render them “nonactionable.” Therefore, given the state of this record, we need not wade into the debate over the wisdom of adopting the incremental harm doctrine, see Kite, supra at 548-63, because even were we to adopt it, it would not apply under the current posture of this case.
B. Issue-Specific Libel-Proof Plaintiffs
The second version of the libel-proof plaintiff doctrine — and the one apparently applied by the trial court in granting summary judgment to the defendants — is known as the issue-specific version. Under it, “[a] libel-proof plaintiff is one whose reputation on the matter in issue is so diminished that, at the time of an otherwise libelous publication, it could not be further damaged.” McBride v. New Braunfels Herald-Zeitung, 894 S.W.2d 6, 9 (Tex. App. 1994); see also Church of Scientology Intern. v. Time Warner, Inc., 932 F. Supp. 589, 593 (S.D.N.Y. 1996) (explaining that “when a particular plaintiff’s reputation for a particular trait is sufficiently bad, further statements regarding that trait, even if false and made with malice, are not actionable because, as a matter of law, the plaintiff cannot be damaged in his reputation as to that trait”). The issue-specific version is typically “applied to justify dismissal of defamation actions where the substantial criminal record of a libel plaintiff shows as a matter of law that he would be unable to recover other than nominal damages.” Jackson v. Longcope, 476 N.E.2d 617, 619 (Mass. 1985). Thus, “[w]hen invoked, the [issue-specific version of the] libel-proof plaintiff doctrine bars a plaintiff from presenting his claim of libel to a jury.” McBride, 894 S.W.2d at 9. Many courts that have adopted the issue-specific version of the doctrine have held that it applies only in a narrow class of cases, “since few plaintiffs will have so bad a reputation that they are not entitled to obtain redress for defamatory statements____” Guccione v. Hustler Magazine, Inc., 800 F.2d 298, 303 (2d Cir. 1986) (citation omitted); see also McBride, 894 S.W.2d at 10.
The United States Court of Appeals for the D.C. Circuit, in an opinion by then Circuit Judge Scalia, rejected the issue-specific version of the doctrine as a “fundamentally bad idea.” Liberty Lobby, Inc. v. Anderson, 746 F.2d 1563, 1569 (D.C. Cir. 1984), reversed on other grounds by Anderson v. Liberty Lobby, Inc., 477 U.S. 242 (1986). However, Liberty Lobby appears to represent a minority view. See Annotation, Defamation: Who Is “Libel-Proof”, 50 A.L.R.4TH 1257 (1986 & Supp. 2006). In fact, we have not been directed to any other case in which a court has reached the same conclusion. While the plaintiff contends that the United States Court of Appeals for the Ninth Circuit rejected the libel-proof plaintiff doctrine, in its entirety, in Masson v. The New Yorker Magazine, Inc., 960 F.2d 896 (9th Cir. 1992), we do not share his reading of that case. In Masson, the *325Ninth Circuit rejected the incremental harm doctrine as “not an element of California libel law.” Id. at 899. The Masson court did not take any position on the issue-specific version of the libel-proof plaintiff doctrine.
Having outlined the nature of the issue-specific version of the doctrine and the debate surrounding it, we reach the question before us: whether to adopt it in this jurisdiction. Like the Massachusetts Supreme Judicial Court, we “accept the principle that a convicted criminal may have such a poor reputation that no further damage to it [i]s possible at the time of an otherwise libelous publication ....” Jackson, 476 N.E.2d at 620; see also Wynberg v. National Enquirer, Inc., 564 F. Supp. 924, 928 (C.D. Cal. 1982). However, we are especially mindful of the Liberty Lobby court’s apprehension about the difficulty courts will face in determining that a reputation has been irreparably damaged. Liberty Lobby, 746 F.2d at 1568. Therefore, while we now adopt this version of the libel-proof plaintiff doctrine, we warn that it should be applied with caution and sparingly.
We hold that:
To justify applying the doctrine, the evidence of record must show not only that the plaintiff engaged in criminal or anti-social behavior in the past, but also that his activities were widely reported to the public. The evidence on the nature of the conduct, the number of offenses, and the degree and range of publicity received must make it clear, as a matter of law, that the plaintiff’s reputation could not have suffered from the publication of the false and libelous statement.
McBride, 894 S.W.2d at 10 (citations omitted). Accordingly, criminal convictions, alone, are not enough to justify application of the doctrine. Compare Cofield v. The Advertiser Co., 486 So. 2d 434, 434-35 (Ala. 1986) (applying the doctrine where no publicity attending the conviction is referenced), with Jackson, 476 N.E.2d at 620 (conviction must be accompanied by publicity), and Wynberg, 564 F. Supp. at 928 (conviction must be accompanied by publicity). Bearing these principles in mind, we turn to the question whether the plaintiff in this appeal is libel-proof.
The trial court found that the plaintiff has criminal convictions in New Hampshire, Massachusetts and Texas for offenses ranging from burglary and receiving stolen property to possession of a controlled substance, criminal threatening, disorderly conduct, resisting arrest, and driving while intoxicated, among others. The trial court also found that: (1) the plaintiff had admitted to some twenty convictions between 1975 and 1990; (2) “the plaintiff has received little media attention regarding his prior arrests and convictions”; and (3) the plaintiff’s “habitual criminal record in *326three ... states” damaged his reputation decades prior to the publication of the Telegraph article. The trial court then concluded that the plaintiff is libel-proof. In light of the principles articulated above, we disagree.
Publicity is part and parcel of the damage to a reputation necessary to trigger the issue-specific version of the libel-proof plaintiff doctrine. Indeed, it is often the means by which such damage occurs and the most effective evidence of that damage. In other cases where courts have most persuasively applied the doctrine and deemed plaintiffs libel-proof, both the publicity surrounding the crimes and the attendant level of notoriety are quite high. See, e.g., Cardillo, 518 F.2d at 640 (plaintiff the subject of testimony before the Federal Congress); Jackson, 476 N.E.2d at 620 (plaintiff the subject of “scores” of newspaper articles); Ray v. Time, Inc., 452 F. Supp. 618, 622 (W.D. Tenn. 1976) (James Earl Ray’s notoriety rendered him libel-proof). The trial court’s findings indicate that no such publicity is present in this case. Accordingly, the trial court erred in deeming the plaintiff libel-proof and its grant of summary judgment on this ground must be reversed.
Reversal on this ground, however, does not end our analysis. The trial court also ruled upon the other grounds for summary judgment asserted by some or all of the defendants as to some or all of the statements. Because the legal issues raised by those rulings are likely to arise on remand, we address them. See Figlioli v. R.J. Moreau Cos., 151 N.H. 618, 622 (2005).
V. The Fair Report Privilege
The Telegraph defendants moved for summary judgment on the ground that the fair report privilege protected all but one of the statements contained in the article. The trial court disagreed and concluded that only nine statements from the article could be moored in the safe harbor of the fair report privilege. It ruled that the other statements fell outside the protections of the privilege because they constituted “information gathered by the police officers during investigations.”
Both sides now appeal that ruling. The plaintiff contends that the trial court erred in applying the fair report privilege because the statements in the article were: (1) based upon confidential information; (2) inaccurate; and (3) made with malice. The Telegraph defendants cross-appeal, contending that the trial court took an overly narrow view of the fair report privilege. They argue that the trial court should have applied the privilege to essentially all statements made by the officers in their official capacities. We first describe the contours of the privilege, and then address the parties’ contentions.
*327As noted above, a plaintiff establishes defamation by showing that the defendant failed to exercise reasonable care in publishing a false and defamatory statement of fact about the plaintiff to a third party, unless a valid privilege applies to the communication. Pierson, 147 N.H. at 763. One such privilege is the fair report privilege. It is a conditional, not absolute, privilege that “applies to the publication of defamatory matter concerning another in a report of an official action or proceeding or of a meeting open to the public that deals with a matter of public concern ... if the report is accurate and complete or a fair abridgement of the occurrence reported.” Hayes v. Newspapers of N.H., 141 N.H. 464, 466 (1996) (quotation and brackets omitted); Restatement (Second) of TORTS § 611 comment a at 297 (1977). A defendant who asserts the fair report privilege bears the burden of establishing its applicability, and the determination of whether the defendant has carried this burden is for the trial court. Id. The privilege extends to the report of any official proceeding, or any action taken by any officer or agency of the government. Id. A report need not track or duplicate official statements to qualify for the privilege; rather, it need give only a “rough-and-ready” summary that is substantially correct. Id. (quotation omitted). In other words, “[a] statement is considered a fair report if its ‘gist’ or ‘sting’ is true, that is, if it produces the same effect on the mind of the recipient which the precise truth would have produced.” Yohe v. Nugent, 321 F.3d 35, 43 (1st Cir. 2003) (quotation omitted). If the privilege does not apply— that is, if a report is not a fair and accurate account of an official proceeding — general fault standards will govern. Hayes, 141 N.H. at 466.
For his part, the plaintiff attempts to convince us that the fair report privilege does not apply because the information contained in the article was based upon confidential materials, such as an allegedly inaccurate presentence investigation report which he claims never to have seen. We are not persuaded. A dispute over the applicability of the fair report privilege is not a viable context in which the plaintiff may litigate or collaterally attack the facts contained in the presentence investigation report. See Yohe, 321 F.3d at 44 (a plaintiff cannot evade the protections of the fair report privilege by merely re-labeling his claim). For the fair report privilege to apply, the article must be an accurate and complete or fair abridgement of the official action or proceeding. Hayes, 141 N.H. at 466. The inquiry does not focus upon truth about the events that either underlie or are the subject of the official action or proceeding. Yohe, 321 F.3d at 44. In other words, “accuracy for fair report purposes refers only to the factual correctness of the events reported and not to the truth about the events that actually transpired.” Id. (quotation omitted).
*328Furthermore, the plaintiff cites no legal authority for the proposition that the information contained in the presentence report is outside the ambit of the fair report privilege. Other courts have expressly rejected such an argument. See Wilson v. Slatalla, 970 F. Supp. 405, 419 (E.D. Pa. 1997) (“the public scrutiny rationale and the public’s interest in the sentencing of convicted criminals support the application of the privilege to the presentence report”). Especially since the presentence report was a part of the official sentencing proceeding, the reasoning from Wilson is persuasive and we adopt it. See RESTATEMENT (SECOND) OF TORTS, supra comments d at 299 and h at 301; see also Yohe, 321 F.3d at 43 (“The fair report privilege protects published reports of arrests by police.”). Therefore, we reject the plaintiff’s argument.
The plaintiff also argues that summary judgment should have been denied because he “submitted affirmative circumstantial evidence to the trial court from which actual malice could be inferred.” The plaintiff’s allegations of “actual malice” appear to encompass claims of ill will towards him and claims that the defendants knew their statements to be false. The Telegraph defendants counter that the fair report privilege “does not depend on the state of mind of the publisher or his belief in the accuracy of the statement reported.” By referencing both state of mind and belief in accuracy, the Telegraph defendants appear to acknowledge the plaintiff’s broad conception of “actual malice” and to argue that neither ill will nor knowledge of falsity can defeat the privilege.
Since the Telegraph defendants’ counterargument raises a threshold legal issue, we address it first. There are two types of malice: (1) constitutional or “actual” malice; and (2) common law malice. Constitutional or actual “malice” is a subjective awareness of the falsity or probable falsity of a statement. Herbert v. Lando, 441 U.S. 153, 156 (1979); New York Times Co. v. Sullivan, 376 U.S. 254, 279-80 (1964) (actual malice is knowledge or reckless disregard of falsity). Common law malice, on the other hand, is ill will or intent to harm. Cantrell v. Forest City Publ’g Co., 419 U.S. 245, 252 (1974). As one court explained, actual malice is concerned with the publisher’s attitude toward the truth while common law malice is concerned with its attitude toward the plaintiff. Geyer v. Steinbronn, 506 A.2d 901, 915 (Pa. Super. Ct. 1986). The plaintiff’s allegations appear to encompass both forms.
The only case the Telegraph defendants cite in support of their position is Yoke. Yoke, however, supports only half of their position. It directly contradicts the other half. In Yohe, the United States Court of Appeals for the First Circuit noted that “the fair report privilege should not be forfeited even if the party making the report knew the statement to be *329false.” Yohe, 321 F.3d at 44 (quotation omitted). The court also noted that “[t]o defeat the [fair report] privilege, a plaintiff must either show that the publisher does not give a fair and accurate report of the official statement, or malice.” Id. (emphasis added). Taken together, these two statements indicate that actual malice cannot defeat the fair report privilege, but common law malice can.
We acknowledge that in some jurisdictions, neither actual nor common law malice can defeat the fair report privilege, see Solaia Technology v. Specialty Pub. Co., 852 N.E.2d 825, 843 (Ill. 2006); Moreno v. Crookston Times Printing Co., 610 N.W.2d 321, 333 (Minn. 2000); however, in other jurisdictions, common law malice can defeat the privilege, a position consistent with the two quoted sentences from Yohe. See, e.g., DeMary v. LaTrobe Printing and Pub. Co., 762 A.2d 758, 764 (Pa. Super. Ct. 2000), appeal denied, 786 A.2d 988 (Pa. 2001); Russell v. Thomson Newspapers, Inc., 842 P.2d 896, 901 n.12, 904-05 (Utah 1992) (summarizing statutory codification of fair report privilege and noting that common law malice can defeat it). We believe that Yoke represents the better approach.
As noted above, the fair report privilege is not an absolute privilege; instead, it is a conditional one, albeit arguably broader than some other conditional privileges. Restatement (Second) of Torts, supra comment a at 297. We have held that conditional privileges can be defeated by a showing of malice involving ill will or intent to harm. See, e.g., Duchesnaye v. Munro Enterprises, Inc., 125 N.H. 244, 253 (1984); Chagnon v. Union-Leader Co., 103 N.H. 426, 438 (1961), cert. denied, 369 U.S. 830 (1962); Huskie v. Griffin, 75 N.H. 345, 348 (1909). These holdings are persuasive, if not controlling.
We are not of the view that allowing a plaintiff to try to establish common law malice as a means of defeating the fair report privilege will unduly chill the free flow of information to the public, especially since it will likely be no easy task for a plaintiff to establish ill will targeted specifically at him or her. See RESTATEMENT (SECOND) OF TORTS, supra Reporter’s Note at 134 (acknowledging that the Restatement (First) of Torts stated that the fair report privilege could be defeated by showing that the publication was “made solely for the purpose of causing harm to the person defamed,” but noting that “there appear to be no cases in which the privilege was lost because of the purpose to harm” (quotation omitted)). Allowing plaintiffs to try to establish common law malice, where appropriate, will guard against abuse of the privilege and ensure that the privilege continues to be used as a shield, not a sword.
To the extent the plaintiff argues that summary judgment should have been denied as to all statements because he submitted *330“circumstantial evidence” of common law malice to the trial court, we reject his argument. The circumstantial evidence cited by the plaintiff is contained in his objection to the defendants’ motions for summary judgment. We have reviewed the plaintiff’s objection and the materials cited therein. They contain, for example, a motion to suppress in a criminal case in which the plaintiff asserted police misconduct. Essentially, the plaintiff argues that summary judgment should have been denied because he feels that police officers acted maliciously toward him. However, that is not the proper inquiry. In the context of the fair report privilege, the malice inquiry — to the extent it is even properly before a court — focuses upon the attitude of the defendant publisher vis-a-vis the plaintiff. See Solaia, 852 N.E.2d at 842-43 (rejecting malice as part of fair report inquiry but thoroughly summarizing and explaining the law in this area). The plaintiff cannot defeat summary judgment wholesale on the fair report privilege by first asserting malice on the part of police officers and then attempting to impute that malice to the Telegraph defendants' by conclusory assertion. See Pennichuck Corp. v. City of Nashua, 152 N.H. 729, 739 (2005) (conclusory allegations not enough to defeat summary judgment). Nor do allegations that the Telegraph was careless amount to malice. Arguments about the accuracy of the article bear upon whether it is accurate and complete or a fair abridgement of the occurrence reported. See Hayes, 141 N.H. at 466. Accordingly, we reject the plaintiff’s argument.
In their cross-appeal, the defendants make essentially two contentions. First, they contend that the trial court took an overly narrow view of the fair report privilege. They argue that the fair report privilege should apply not only to all “rough and ready” summaries of official records and documents, but also to news reports of any “oral statements made by law enforcement officers and other public officials in their official capacity.” Second, they contend that even under a narrower construction of the privilege, the trial court still erred in failing to grant summary judgment as to certain statements.
With respect to the first contention, we agree that the privilege should apply to all public, official actions or proceedings. However, the Telegraph defendants ask us to decide an issue broader than the one before us. We need not decide the extent to which all oral statements by both law enforcement officers and other public officials would fall within the privilege because only statements by police officers, imputing criminality, are here at issue. Therefore, our analysis is narrow.
As to the police officers’ statements, we acknowledge that the fair report privilege has been construed broadly in some jurisdictions to cover essentially any communication to a reporter by a law enforcement officer. *331See generally Annotation, Defamation: Privilege Attaching to News Report, of Criminal Activities Based on Information Supplied by Public Safety Officers — Modem Status, 47 A.L.R.4th 718-39 (1986 & Supp. 2006).
In other jurisdictions, the privilege has been construed more narrowly:
[T]he fair report privilege should not be extended to apply to the myriad types of informal reports and official and unofficial investigations, contacts, and communications of law enforcement personnel at all levels of the state and federal bureaucracy with local, regional, and national media____
The American Law Institute has addressed how far the scope of “official action” extends into the domain of arrests and the underlying facts associated with the arrests by differentiating between reports of an arrest and statements regarding the underlying facts that precipitated the arrest. The Restatement (Second) of Torts states that an airest by an officer is an official action, and a report of the fact of the arrest or of the charge of crime made by the officer in making or returning the arrest is within the fair report privilege; however, statements made by the police or by the complainant or other witnesses or by the prosecuting attorney as to the facts of the case or the evidence expected to be given are not yet part of the judicial proceeding or of the arrest itself____
Lewis v. NewsChannel 5 Network, L.P., No. M2005-00458-COA-R3-CV, 2007 WL 1585163, at *13-*14 (Tenn. Ct. App. May 31, 2007) (quotations, citations and brackets omitted).
Our view of the privilege lies between these two approaches. See D. Elder, The Fair Report Privilege §§1.08-1.10, at 77-110 (1988) (explaining how jurisdictions apply the fair report privilege to law enforcement activities and criminal prosecutions). We do not view all private conversations between police officers and reporters to be automatically within the privilege. For example, “some unofficial version of events furnished by a policeman at a crime scene, or ... offhand prediction” would not be within the privilege. Id. § 1.08, at 77 (quotations and brackets omitted). These types of situations do not support a contention that the media is justified in using the privilege as a defense because it is functioning as “the eyes and ears of the public.” Costello v. Ocean County Observer, 643 A.2d 1012, 1020 (N.J. 1994).
In Hayes, we held that the fair report privilege applies to “[t]he publication of defamatory matter concerning another in a report of an official action or proceeding or of a meeting open to the public ....” Hayes, 141 N.H. at 466 (quotation omitted and emphases added). Thus, it *332is the terms “official,” “action,” “proceeding” and “meeting” that are critical to the application of the privilege in this state. Because not every conversation will involve “an official action or proceeding or ... a meeting open to the public,” id. (quotation omitted), it would constitute a marked expansion of Hayes if we were to adopt the broad view of the privilege taken in some jurisdictions and hold that the fair report privilege applies to all conversations between law enforcement personnel and reporters.
On the other hand, a narrow approach like the one adopted in Lems does not account for all types of “conversations” that may occur between a reporter and a police officer, and would fall within the privilege. The privilege also protects reports that meet the accuracy requirements of Hayes, and are based upon press conferences, interviews with a police chief, see Yohe, 321 F.3d at 44, or other types of official “conversations.” Cf, e.g., Restatement (Second) of Torts, supra comment d at 299.
The information at issue here came from private conversations between Trudell and the officers. The article demonstrates that the conversations went well beyond the fact of the plaintiff’s arrest and the grounds for the charge. Indeed, the officers’ comments reference related investigations and the plaintiff’s prior criminal history.
We have been directed to no evidence that the officers were given the official imprimatur of their departments to function as spokesmen or even to speak with Trudell. The officers’ affidavits state only that the shared information “included” matters of public record. Therefore, we have no way of assessing whether all or any part of the investigations discussed were public at that time. Certainly, a police department does not make public all aspects of all investigations. Doing so would seriously undermine its ability to combat crime. Thus, for all that appears in the record the defendants in this case were police sources making unofficial statements.
It bears noting that the Hudson Police Department, for example, operates under a media relations policy that prohibits individual officers from revealing to reporters, without the approval of the police chief, “[t]he prior criminal history of the accused,” “[statements regarding the character or reputation of the accused,” and “[t]he evidence in the case.” We do not take a position on whether and the extent to which this policy is consistent with the fair report privilege. Those questions are not before us. Instead, we cite the Hudson policy to illustrate that our holding today does not upset the practices of at least some police departments in this state, including one whose officers are involved with this case.
We now turn to the Telegraph defendants’ second contention concerning specific statements that they believe fall within the privilege. The trial court found that “statements 9,10, 24, 27, 30, 51, 52, 53, and 54 ... are fair and accurate reports of the contents or records of official actions,” but that *333the other “statements [in the article] consisted of information gathered by the police officers during investigations” and were therefore not covered by the fair report privilege. Because we hereafter uphold the trial court’s grant of summary judgment as to statements 51-54 on the basis of substantial truth, we take no position on the application of the fair report privilege to those statements.
The Telegraph defendants argue that all of the statements attributable to the officers fall within the privilege because the officers were summarizing the contents of police records. With respect to the records Gosselin allegedly summarized, the Telegraph defendants cite pages 103 to 120 of the appendix, all but two of which are documents dated after the publication of the article. Consequently, on this record, we have no way of evaluating whether Gosselin was relying upon or summarizing any given record at the time of his interview. Other police records are also contained in the appendix, but neither the Telegraph defendants nor the officers’ affidavits explain which particular records the officers relied upon at the time of the interview. Thus, even assuming the Telegraph defendants’ argument is legally correct, the absence of such an explanation and the requirement that all inferences must be construed in favor of the plaintiff mean that summary judgment was correctly denied.
The Telegraph defendants argue that statements 20-23 are also within the privilege because they “were derived from police and court records, [which] are entitled to the protection of the privilege.” Again, we emphasize that our discussion in this case is limited to police records. We agree that official police records, such as official blotters, official reports, and so forth, fall within the privilege. Elder, supra § 1.08, at 77; see also Restatement (Second) of Torts, supra, comment d at 299 (“The filing of a report by an officer ... of the government is an action bringing a reporting of the governmental report within the scope of the privilege.”); cf. Costello, 643 A.2d at 1021 (a complaint not yet filed does not constitute an official record and therefore is not subject to the fair report privilege); Wynn v. Smith, 16 P.3d 424, 430 (Nev. 2001) (“unauthorized or confidential investigatory [police] reports do not qualify as an ‘official action or proceeding’ under the fair report privilege”).
In support of their position that statements 20-23 derive from official records, the Telegraph defendants cite six pages of the appendix. All of these pages are copies of a typewritten facsimile transmission. At the top of one page is printed “Seabrook Police Department Arrest Report” and at the top of another is printed “Seabrook Police Department Initial Investigation Report.” None of the pages is signed, not even where the document entitled “Initial Investigation Report” contains lines requiring signatures from the investigating officer and the reviewing officer. In *334addition, none of the six pages is accompanied by an affidavit indicating that it is “official.” Nor does the Telegraph’s brief direct us to any part of the record that would establish that these unsigned documents are “official” or in the public record. Other police records in the appendix are .signed.
Documents authored by police officers do not become “official” or matters of public record simply because they may be located in the police department. Nor do they become “official” or matters of public record simply because they are consistent with what a reporter might have heard while conversing with people for a news story. They become “official” when they bear adequate indicia of being “official” or are actually in the public record. In addition to other circumstances, documents may be deemed “official” when they are signed, correctly executed, filed or accompanied by an affidavit indicating they are official. Cf. RESTATEMENT (SECOND) OF Torts, supra comments d at 299 and h at 301; Elder, supra § 1.10, at 87-110. On this record, we have no way of knowing anything about the nature of these unsigned reports. It is the defendants’ responsibility, as appellants on this issue, to provide an adequate record to support their assertions. See SUP. Ct. R. 13(2), 16(3)(d).
Further, even if we were to assume that the “Arrest Report” is “official,” none of the information in statements 20-23 comes from that page. Nor does that page expressly incorporate the subsequent five. Thus, we reject the Telegraph defendants’ arguments as to statements 20-23 and uphold the trial court’s decisions regarding summary judgment, albeit on somewhat different grounds.
VI. Substantial Truth
All of the defendants moved for summary judgment on the basis that the statements in the article are substantially true. The trial court found that only some of the statements were substantially true, while the other statements “primarily deal[t] with the defendant police officers’ suspicions of the plaintiff committing numerous crimes.”
On appeal, the parties’ arguments on the issue of substantial truth are essentially two-fold. First, both sides contend that the trial court’s analysis of substantial truth was incorrect because the court considered individual statements rather than the article as a whole. The plaintiff contends that the gist or sting of the entire article is defamatory, while the defendants argue that: (1) the gist or sting of the article, read as a whole, is substantially true; and (2) each defendant’s comments, considered in the aggregate, constitute a “statement,” and the “statement” of each defendant is substantially true.
*335Second, the parties dispute the trial court’s rulings on the substantial truth of particular statements. The plaintiff argues that summary judgment should have been denied as to each statement because each either is untrue or misrepresents the facts. The defendants, on the other hand, urge us to affirm the trial court’s decision that nine of the statements from the article are substantially true. We start with the parties’ broad attack on the trial court’s method of analysis, and then consider their arguments concerning the trial court’s rulings on specific statements.
One who publishes a defamatory statement of fact is not subject to liability for defamation if the statement is true. In the law of defamation, truth is defined as substantial truth, as it is not necessary that every detail be accurate. In other words, literal truth of a statement is not required so long as the imputation is substantially true so as to justify the gist or sting of the remark. Furthermore, a false and defamatory inference may be derived from a factually accurate news report.
Faigin v. Kelly, 978 F. Supp. 420, 425 (D.N.H. 1997) (quotations omitted); see also RESTATEMENT (SECOND) OF TORTS § 581A at 235-37. While substantial truth is a defense to a libel or defamation claim, such a
claim can only rarely be dismissed on the rationale that the statements complained of are substantially true, as the notion of substantial truth necessarily implies a thread of untruth, and the conclusion that a statement is substantially true will therefore involve a determination that whatever errors are in the statement are irrelevant in the minds of the audience.
53 C.J.S. Libel and Slander; Injurious Falsehood § 164 (2005).
As noted above, both the plaintiff and the defendants contend that the trial court erred in its method of evaluating substantial truth because it evaluated the substantial truth of each challenged sentence individually. The parties offer two alternative methods of analysis that they believe should have been followed.
Gosselin contends that the trial court’s analysis of substantial truth is deficient because the court misapprehended the term “statement.” The trial court treated each challenged sentence as a statement and evaluated each of those in the context of the whole article. However, in Gosselin’s view, a statement, for purposes of substantial truth, is defined as the aggregate of all sentences attributable to an individual speaker. Gosselin does not cite any legal authority in support of this position. We decline *336Gosselin’s invitation to define a statement as an aggregate of sentences attributable to individuals. We see little merit in a rule that would allow a defendant to avoid liability by simply couching injurious and baseless sentences in a longer “statement.” The better view is that the “statement” giving rise to liability can be one of an individual’s remarks or many, while the interview as a whole provides important context for evaluating whether the “statement” is substantially true or an opinion. Accordingly, we reject Gosselin’s argument and turn to the second method of analyzing substantial truth advanced by the parties.
Both the plaintiff and the defendants argue that the trial court should have evaluated the substantial truth of the article as a whole rather than the substantial truth of each challenged statement individually. While we agree that courts must consider defamatory statements in context, we discern no error in the trial court’s approach here. Courts analyze claims of substantial truth by examining individual sentences within a news report, see, e.g., John v. Journal Communications, Inc., 801 F. Supp. 199, 200-10 (E.D.Wis. 1992); Foretich v. CBS, Inc., 619 A.2d 48, 63 n.22 (D.C. 1993), in light of the context of the report as a whole, see, e.g., 50 Am. Jur. 2d Libel and Slander § 251 (2006); 53 C.J.S. Libel and Slander; Injurious Falsehood § 34 (2005) (“Allegedly defamatory words are to be considered in context under the circumstances of their publication, as the context of a statement can significantly affect its fair and natural meaning.”). This approach is consistent with our case law discussing how statements of opinion must be evaluated in the defamation context. See Nash v. Keene Publishing Corp., 127 N.H. 214, 219 (1985).
The plaintiff challenged approximately fifty-eight statements from the article. Thus, each challenged and actionable statement had to be evaluated in the context of the article as a whole. The trial court’s order demonstrates that this is what it did: it identified the precise statements that it deemed substantially true, mindful of the context of the article as a whole. Accordingly, we uphold the trial court’s method of analysis and turn to the parties’ arguments concerning whether the trial court properly applied that analysis.
The plaintiff argues that the trial court erred in entering summary judgment on the issue of substantial truth because the statements at issue are inaccurate or misrepresent the facts. Four of the statements the plaintiff contends are inaccurate pertain to a conviction or sentence the plaintiff received in connection with criminal activity. “If the defamatory statement is a specific allegation of the commission of a particular crime, the statement is [deemed] true [for purposes of a substantial truth defense] if the plaintiff did commit that crime.” RESTATEMENT (SECOND) *337OF TORTS § 581A comment c at 236. Statements 9, 24, 27 and 30 fall within this rule.
Next, the trial court found that statements 10, 51, 52, 53 and 54 “are based upon official records and proceedings.” The plaintiff nevertheless contends that they are not substantially true. The plaintiff’s brief does not point to any portion of the record creating a genuine issue of material fact with respect to statement 10. See SUP. CT. R. 16(3)(d). More significantly, the basis for statement 10 was deemed admitted by an order of the trial court that has not been appealed. Therefore, we uphold the trial court’s conclusion regarding statement 10 as well.
Statements 51-54 pertain to the circumstances surrounding the plaintiff’s arrest in 1999. They report that the plaintiff was in possession of a bag containing stolen property when he was arrested. However, the plaintiff produced, in an appendix to his objection to the motions for summary judgment, testimony from one of the arresting officers indicating that this bag actually was found in a dumpster near where he was arrested. This factual discrepancy is significant for purposes of summary judgment here only if it is material to the determination of substantial truth. See Coco, 154 N.H. at 356 (defining “material”). We conclude that it is not material. The location of the stolen property and the plaintiff’s possession of it may be considered important facts in the prosecution of the plaintiff; however, the gist or sting of these statements is that the plaintiff was arrested for receiving stolen property in Hudson. Whether or not he possessed the stolen items at the time of his arrest does not bear upon whether the fact of his arrest for that particular crime is substantially true. See Faigin, 978 F. Supp. at 425 (disputed issue of fact as to whether gist or sting is substantially true). Accordingly, we uphold the trial court’s grant of summary judgment as to statements 51, 52, 53 and 54 on the basis of substantial truth.
As for the defendants, they appear to contend that the statements on which the trial court did not grant summary judgment are substantially true because they pertain to criminal activity, the plaintiff admitted a number of previous criminal activities, and his involvement in criminal activity is clear from the article when it is read as a whole. While the plaintiff may have been involved in criminal activity, he denies the facts underlying the challenged statements at issue here, thereby rendering substantial truth a question for the jury. For example, one of the statements indicates that the plaintiff is suspected of having perpetrated “more than 1000 home burglaries.” There can be no doubt that publishing suspicions about the plaintiff’s involvement in additional crimes (beyond those to which he admitted or for which he has been arrested), even in the *338context of the whole article, is capable of defaming him. See Catalfo v. Jensen, 657 F. Supp. 463, 466 (D.N.H. 1987) (“Whether a communication is capable of bearing a defamatory meaning is an issue of law [for] the Court. Only if the Court determines that language is defamatory is there then the question for the jury whether the communication was in fact understood by its recipient in the defamatory sense.”); see also Burke v. Town of Walpole, 405 F.3d 66, 94-95 (1st Cir. 2005) (‘Words may be found to be defamatory if they hold the plaintiff up to contempt, hatred, scorn or ridicule, or tend to impair his standing in the community. Imputations of criminality generally fit the bill.” (quotation and citations omitted)); Chagnon v. Union-Leader Co., 103 N.H. 426, 434 (1961). And, while it may be true that the plaintiff was a suspect in a number of burglaries, it remains for the jury to determine the substantial truth of the defendants’ characterizations of those suspicions, especially since the plaintiff denies them. The same reasoning applies to the other statements with regard to which the trial court declined to enter summary judgment.
Accordingly,
[although it is not necessary for the defendants] to prove the literal truth of a defamatory statement in every detail but only that it is substantially true[,] we cannot say as a matter of law that the defendants] met that burden in this case. The jury could find on the evidence that the defamatory statements ... were untrue.
Chagnon, 103 N.H. at 437. Therefore, based upon our review of the record, we uphold the trial court’s denial of summary judgment as to the remaining statements.
VII. Statements of Opinion
All of the defendants moved for summary judgment on the basis that their respective statements were not libelous because they were statements of opinion. The trial court granted summary judgment with respect to Flynn’s statements, concluding that they were statements of opinion, but denied summary judgment with respect to the police defendants’ statements, concluding that they were statements of alleged facts or implied, undisclosed defamatory facts.
A statement of opinion is not actionable unless it may reasonably be understood to imply the existence of defamatory fact as the basis for the opinion. Nash, 127 N.H. at 219; see also Milkovich v. Lorain Journal Co., 497 U.S. 1, 18-19 (1990). Whether a given statement can be read as being or implying an actionable statement of fact is a question of law to be *339determined by the trial court in the first instance, considering the context of the publication as a whole. Nash, 127 N.H. at 219. If an average reader could reasonably understand a statement as actionably factual, then there is an issue for a jury’s determination and summary judgment must be denied. Id.
Citing two affidavits attached to his objection to the motions for summary judgment, the plaintiff argues that all of Flynn’s statements could be understood as asserting facts or as being based upon undisclosed facts. Not so. As the trial court noted, Flynn’s statements constituted her
opinion about the plaintiff and [were] based completely on information provided by others as disclosed in the article. It cannot reasonably be concluded that Flynn’s opinion was based on any undisclosed facts, defamatory or otherwise. She was simply presented with a set of hypothetical facts which were disclosed in the article and rendered an opinion limited to those facts.
The affidavits cited by the plaintiff merely summarize passages from the article, and contain assertions that the affiants could not imagine the plaintiff would commit the criminal activity referenced in the article. Such assertions do not defeat summary judgment as to Flynn. See Salitan v. Tinkham, 103 N.H. 100, 103 (1960) (affidavits must contain facts not legal conclusions). Accordingly, we uphold the trial court’s grant of summary judgment to Flynn and to the Telegraph defendants as to Flynn’s statements.
Gosselin, Anderson, Droney and Bousquet contend that their statements were also protected opinions. However, we agree with the trial court that even assuming arguend,o these statements are opinions, they are clearly based upon undisclosed facts resulting from unspecified investigations. See Jorg v. Cincinnati Black United Front, 792 N.E.2d 781, 784 (Ohio Ct. App. 2003) (accusations of criminal activity generally give rise to “clear factual implications”). Thus, an average reader could reasonably understand the officers’ statements as actionably factual, and summary judgment was properly denied as to them on this ground. Nash, 127 N.H. at 219.
The cases cited by the defendants do not change our conclusion. For example, in Jorg, a police officer brought a defamation action against a civil rights organization that distributed a letter to local media, which was published, generally accusing the police department and some of its members of murder, rape and planting false evidence. Jorg, 792 N.E.2d at 782-83. The Ohio Court of Appeals noted that oftentimes these types of *340accusations would be considered factual — as opposed to opinionative. Id. at 784-85. However, given the unique nature and tenor of the letter, the court concluded that it was “hyperbole,” “meant to be persuasive,” “a call to action,” and “meant to cause outrage in the reader.” Id. at 786. Indeed, the court deemed the letter at issue “a persuasive piece of advocacy and not a news article purporting to be objective reporting.” Id. As such, the letter and its statements were very clearly opinionative. In contrast, the article at issue in this appeal is not intended to be a persuasive call to action. Rather, it is presented as an objective reporting of the plaintiff’s alleged and admitted criminal activities. Accordingly, we conclude that Jorg is distinguishable.
VIII. Limited Purpose Public Figure and Actual Malice
The police defendants contended that they were entitled to summary judgment because the plaintiff should be considered a limited-purpose public figure who cannot prove actual malice. The trial court found “that the plaintiff is a private person and not a public figure and thus, is not required to prove that the statements were made with ‘actual malice.’ ”
In an effort to strike a balance between First Amendment freedoms and state defamation laws, [we] accord[] ... significance to the [public or private] status of each individual plaintiff. Under the taxonomy developed by the [United States] Supreme Court, private plaintiffs can succeed in defamation actions on a state-set standard of proof (typically, negligence), whereas the Constitution imposes a higher hurdle for public figures and requires them to prove actual malice.
Pendleton v. City of Haverhill, 156 F.3d 57, 66 (1st Cir. 1998); see also Gertz v. Robert Welch, Inc., 418 U.S. 323, 347 (1974). Determining whether an individual is a public or private figure presents a threshold question of law, see Nash, 127 N.H. at 222, which is “grist for the court’s — not the jury’s — mill.” Pendleton, 156 F.3d at 67.
The United States Supreme Court has created two subclassifications of public figures: (1) persons who are public figures for all purposes; and (2) so-called limited-purpose public figures who are public figures for particular public controversies. Gertz, 418 U.S. at 351. With respect to the first group, “an individual may achieve such pervasive fame or notoriety that he becomes a public figure for all purposes and in all contexts.” Id. When determining that an individual is this type of public figure, courts should
*341not lightly assume that a citizen’s participation in community and professional affairs rendered him a public figure for all purposes. Absent clear evidence of general fame or notoriety in the community, and pervasive involvement in the affairs of society, an individual should not be deemed a public personality for all aspects of his life.
Id. at 352. In addition and significant to this appeal, a person does not automatically become this type of public figure simply because he engages in criminal activity. Wolston v. Reader’s Digest Assn., Inc., 443 U.S. 157, 167 (1979); see also Bender v. City of Seattle, 664 P.2d 492, 504 (Wash. 1983) (“[A] person is not considered a ‘public figure’ solely because that person is a criminal defendant, has sought relief through the courts, or is involved in a controversy which is newsworthy.” (citations omitted)).
As to the second group, individuals may become limited-purpose public figures when they “have thrust themselves to the forefront of particular public controversies in order to influence the resolution of the issues involved.” Gertz, 418 U.S. at 345. Then, they “become[] a public figure for a limited range of issues.” Id. at 351. Courts make the limited-purpose public figure determination “by looking to the nature and extent of an individual’s participation in the particular controversy giving rise to the defamation.” Id. at 352.
Finally, we must draw a distinction between these public figures and private citizens.
Even if the foregoing generalities do not obtain in every instance, the communications media are entitled to act on the assumption that public officials and public figures have voluntarily exposed themselves to increased risk of injury from defamatory falsehood concerning them. No such assumption is justified with respect to a private individual. He has not accepted public office or assumed an influential role in ordering society. He has relinquished no part of his interest in the protection of his own good name, and consequently he has a more compelling call on the courts for redress of injury inflicted by defamatory falsehood. Thus, private individuals are not only more vulnerable to injury than public officials and public figures; they are also more deserving of recovery.
Id. at 345 (quotations and citation omitted). Accordingly, private plaintiffs need not establish actual malice to recover actual damages. See Pendleton, 156 F.3d at 66.
*342We now apply these general principles to the present case. The defendants argue that the plaintiff is a limited-purpose public figure by virtue of his involvement in numerous criminal activities. Although a person does not automatically become an all-purpose public figure simply because he engages in criminal activity, some courts have ruled that a convicted criminal can become a limited-purpose public figure depending upon the nature of his involvement in crime. See, e.g., Ruebke v. Globe Communications Corp., 738 P.2d 1246, 1252 (Kan. 1987). For example, in Ruebke, the plaintiff had committed a triple murder, the resulting investigation of which was highly publicized and thrust him to the forefront of public attention. Id. The court held that the crime committed was a public controversy and that the nature and extent of Ruebke’s participation in it was enough to cause him to become a public figure. Id. Similarly, in Talley v. WHIO TV-7, 722 N.E.2d 103, 106-07 (Ohio Ct. App. 1998), the court held that a defendant was a limited-purpose public figure where he attempted to murder his wife by stabbing her repeatedly, and the media covered his arrest and trial. The court determined that the defendant’s crime was a matter of substantial public interest and controversy. Id. at 107.
Here, by contrast, the crimes to which the plaintiff has admitted (and those to which he has not) are not matters of public controversy. They are burglaries that, while serious, affected those whose homes had been burgled but did not otherwise garner much public attention. Indeed, as the trial court quite properly found, “It is only because of the article that any publicity as to the plaintiff was generated. There is no independent evidence to suggest that the plaintiff has obtained independent notoriety or special prominence in the public eye.” Accordingly, we uphold the trial court’s conclusion that the plaintiff was not a limited-purpose public figure and therefore would not be required to establish actual malice.
IX. Qualified Privilege
All of the police defendants argued to the trial court that they were entitled to summary judgment because they enjoyed a qualified privilege to make their remarks. They argued that the qualified privilege protected their communications with Trudell because they spoke with him in the course of their official duties as law enforcement agents. The trial court held that “the police officers’ statements were not conditionally privileged.”
On appeal, Gosselin argues that application of the privilege is justified “both as a specific cautionary tale of the residents’ vulnerability to theft and to demonstrate the service provided to the Town by its tax-supported *343police department.” Gosselin further contends that the police had a legitimate interest in educating the public about the facts of the plaintiff’s arrest, their extensive investigation, and the importance of the arrest.
As we noted earlier, application of a privilege can absolve a defendant from liability for libelous statements. Pierson, 147 N.H. at 763-64. Privileged communications are generally divided into two classes: (1) those that are absolutely privileged; and (2) those that are qualifiedly or conditionally privileged. Id. at 764. On one hand, if a communication is absolutely privileged, the speaker is absolutely immune from suit regardless of his or her motive in making the communication. Id. at 763-64. Absolute privileges are present in a narrow class of cases. Id. at 764. On the other hand, a conditional or qualified privilege may be “established if the facts, although untrue, were published on a lawful occasion, in good faith, for a justifiable purpose, and with a belief, founded on reasonable grounds of its truth.” Touma v. St. Mary’s Bank, 142 N.H. 762, 765 (1998). Facts or evidence used in mounting a substantial truth defense may overlap with those used to demonstrate the applicability of a qualified or conditional privilege (i.e., to show good faith), but the two defenses are not the same and the inability to claim one does not necessarily preclude an individual from claiming the other. In any event, if all of the circumstances enumerated above are not present or if there is actual malice, the speaker’s immunity from suit may be lost. Pierson, 147 N.H. at 764; see also Simpkins v. Snow, 139 N.H. 735, 740 (1995).
Since, under our test, a speaker is not always entitled to claim a conditional or qualified privilege, it follows that “[n]ot every statement made to a newspaper reporter by a police officer in the course of an investigation is protected ....” Lee v. City of Rochester, 600 N.Y.S.2d 564, 565 (App. Div. 1993). This is true even though the qualified privilege protects the need “to allow public officials to speak freely on matters of public importance in the exercise of their official duties.” Burke, 405 F.3d at 94.
Other jurisdictions have also concluded that, with respect to police officers, application of the qualified privilege has its limits. For example, the Louisiana Supreme Court held
that law enforcement officers, whose duty includes charging persons with crimes, should be allowed to report the fact of a criminal investigation and an arrest without fear of a defamation action if the person is cleared of the charges, [however,] an officer cannot add additional injurious statements that the officer had no reason to believe were true. Such a restriction of the privilege should not have a chilling effect on the free reporting of *344criminal investigations and arrests, but should prevent occurrences ... where the officer not only reported the investigation and arrest, but also reported facts pertaining to guilt that were not developed in the investigation.
Trentecosta v. Beck, 703 So. 2d 552, 564 (La. 1997). Similarly, the Washington Supreme Court held that “[t]he right to inform the public ... does not include a license to make gratuitous statements concerning the facts of a case or disparaging the character of other parties to an action.” Bender, 664 P.2d at 504.
Relying primarily upon Slocinski v. Radwan, 83 N.H. 501, 504 (1929), the trial court ruled that the officers could not claim the privilege because: (1) the statements were not made to release important information to the media and the general public on an arrest, a possible threat to the community, or law enforcement practices; (2) they were not published to a narrow group that shared an interest in the communication; and (3) the police officers and the Telegraph defendants did not share a corresponding duty or interest in the subject. In so ruling, the trial court appears to have fashioned a more particularized formulation of the qualified privilege inquiry and then ruled, as a matter of law, that the privilege did not apply under that more particularized test.
We disagree with the trial court’s determination for two reasons. First, we acknowledge that some jurisdictions appear to view the speaker’s entitlement to claim a qualified privilege as a question of law, see, e.g., Bender, 644 P.2d at 504-05; Burke, 405 F.3d at 94-95; however, we have held that it is one of fact. See Pierson, 147 N.H. at 764 (“the question whether the speaker is entitled to claim the [qualified] privilege is one for the trier of fact”). Second, we have not previously articulated a special formulation of the qualified privilege in the case of police defendants, and we decline to do so here. Rather, in light of the above-described limitations upon when a speaker is entitled to claim the privilege, we conclude that the test laid out in Pierson, 147 N.H. at 763-64, adequately safeguards the interests of law enforcement personnel and defamation plaintiffs. Consequently, it is this formulation of the privilege that should apply.
Accordingly, we vacate the trial court’s ruling on this issue. See Pierson, 147 N.H. at 764. On remand, the trier of fact should determine the availability of the privilege by deciding whether it is “established if the facts, although untrue, were published on a lawful occasion, in good faith, for a justifiable purpose, and with a belief, founded on reasonable grounds of its truth.” Touma, 142 N.H. at 765. These determinations may require a trial, or, if the court finds no genuine issues of material fact, the applicability of the privilege may be resolved on summary judgment. See *345Concord Group Ins. Co’s v. Sleeper, 135 N.H. 67, 69 (1991) (“It has been recognized that the presence of a question involving state of mind or intent does not automatically foreclose the application of summary judgment, but it should be cautiously and sparingly invoked in such instances.”).
X. Conclusion
Consistent with our analysis above and for the foregoing reasons, we reverse the trial court’s grant of summary judgment on the issue of Thomas being a libel-proof plaintiff and vacate its ruling on the issue of the applicability of a qualified privilege to the police defendants. In all other respects, we affirm.
Affirm,ed in part; reversed in part; vacated in part; and remanded.
Broderick, C.J., and Dalianis, J., concurred.