OPINION OF THE COURT
Titone, J.In 1979 the Newburgh City Council was considering a reorganization of the city’s police department. Debate was focused on the question of whether the city should continue with the existing appointed commissioner system or return to a civil service police chief structure. On July 19, 1979 *227the Times Herald Record voiced the opinion that the commissioner system should be retained. To support his argument the editor commented on past events and past accusations leading to the installation of the commissioner system.
“The department was in a shambles in 1972 after 18 officers, including Chief Humbert Capelli, were indicted on charges of burglary, planting of evidence and other misdeeds.
“We said at the time, and we still believe, that the entire department was under a cloud. It is inconceivable to us that so much misconduct could have taken place without the guilty knowledge of the unindicted members of the department. If so, they all were accessories after the fact, if not before and during.”
Within a month of publication, the owner of the Times Herald Record and certain individuals connected with the paper were sued for libel. The action was instituted by 27 present members of the police force who were among the group that had not been indicted in 1972.
In a preliminary conflict generated by the service of a second set of interrogatories by a police officer other than the one who served the first set, both the officer who served the second set (plaintiff), and the media defendants moved for summary judgment. The cross motions were denied and the parties appeal.
In our opinion, Justice Kelly’s determination was proper. Discovery is not yet complete and the necessary factual clarification of the major issues has not occurred. Therefore we affirm, without prejudice to renewal when the discovery has been completed.
We are forced to go beyond the decision of Special Term, however, because the posture of the case on appeal has changed. The newspaper has raised certain contentions that were not argued at Special Term or were argued in a different form. Specifically, the paper now challenges the reference in the editorial comments to the plaintiff and asserts that the comments are protected by the new opinion privilege. The appealing police officer, while noting the paper’s failure to articulate these claims at Special Term, *228has responded to them. We have chosen to make the necessary determination. The issues are central to all 27 cases and have the potential to terminate the litigation or to refocus the remaining discovery.
REFERENCE TO THE INDIVIDUAL PLAINTIFF IN THE CONTEXT OF GROUP DEFAMATION
The “of and concerning” element in defamation actions requires that the allegedly defamatory comment refer to the plaintiff (<Julian v American Business Consultants, 2 NY2d 1). The media defendants have maintained for the first time on appeal that the plaintiff cannot show that the editorial comments were “of and concerning” him in view of the size of the group to which the editorial refers. In 1972 the unindicted police officers employed by the City of Newburgh apparently numbered at least 53.1 Accordingly, the newspaper contends that reference to the individual police officer cannot be shown as a matter of law because the group allegedly defamed constitutes an indeterminate class exceeding 25 individuals. The plaintiff responds that in this case a group was libeled rather than a class and that therefore he need show only that he was a member of that group in order to demonstrate reference to him.
Generally, an impersonal reproach of an indeterminate class is not actionable (Gross v Cantor, 270 NY 93, 96), for “one might as well defame all mankind” (Prosser, Law of Torts [4th ed], p 750). The underlying premise of this principle is that the larger the collectivity named in the libel, the less likely it is that a reader would understand it to refer to a particular individual (Sack, Libel, Slander, and Related Problems, § II.8.3). A group is a subset of a class, that is, a limited or restricted portion or collection of a class (23 LRA[NS] 726, case note). Nevertheless, since the possibility of specific reference diminishes as the size increases, where the group named approaches class size, no personal reference can be found.
The psychological assumption inherent in this premise is *229that the reader will treat the statement rationally.2 “The theory of the common-law rule against group-libel actions is that the gravamen of such an action must be individual harm, and that as the target group increases in size, the harmful effect of the statement on any individual member must be diluted, until at some point the harm falls below the threshold of legal recognition. To be plausible, this theory requires the positing of a determinedly rationalistic, even mechanical, model of the human psyche. Such a model supposes that the hearer of a group-libelous statement always treats that statement rationally: in other words, that the hearer does not allow the statement to affect his estimation of an individual member of the target group without reasonable grounds to believe that the general statement holds true for the particular individual. Given this supposition, the model leads to a belief that individual harm cannot occur as the result of a group-libelous statement, because the hearer of the statement will make the rational assessment that such a statement is, by its nature, less likely to be true with respect to every member of a large group than it is to be true with respect to a particular individual. The model thus concludes that the hearer will discount the statement according to the size of the target group, and that he will be influenced in his subsequent dealings with individual members of the group only to this fractional, discounted extent.” (Group Vilification Reconsidered, 89 Yale LJ 308, 310-311.)
The rule was designed to encourage frank discussions of matters of public concern under the First Amendment guarantees. Thus the incidental and occasional injury to the individual resulting from the defamation of large groups is balanced against the public’s right to know. (See Service Parking Corp. v Washington Times Co., 92 F2d 502, 505-506; Riesman, Democracy & Defamation: Control of Group Libel, 42 Col L Rev 727.) As early as 1840 New York recognized this conflict with First Amendment values in group defamation in Ryckman v Delavan (25 Wend 186, *230198-199): “It is the malicious intention of the libeller towards the injured individual that authorizes the latter to seek redress. The proof, or else the necessary presumption of individual malice, and the inflicting individual injury, are the sole grounds of the civil action and of the remedy it affords. General censure or reproof, satire or invective, directed against large classes of society, whether on moral, theological or political grounds, cannot ordinarily be prompted by individual malice or intended to produce personal injury. The politician who assails the opposite party, the polemical divine who attacks the doctrine or the discipline of another church or sect, or the moral satirist who lashes the vices or the foibles of his age and nation, ought not to be held responsible in private suits for the bold avowal of opinions true or false. The principle upon which the civil remedy is allowed, does not apply here; and the great interests of society require that it should not be made to apply. It is far better for the public welfare that some occasional consequential injury to an individual, arising from general censure of his profession, his party, or his sect, should go without remedy, than that free discussion on the great questions of politics, or morals, or faith, should be checked by the dread of embittered and boundless litigation.”
In recent times, the principle has had frequent application. In Mansour v Fanning (506 F Supp 186, 187) a class action on behalf of over 600,000,000 Muslims was brought to recover damages for the airing of the film “Death of a Princess”. The plaintiffs claimed that the film, which depicted the public execution of a Saudi Arabian princess for adultery, was insulting and defamatory to the Islamic religion. In dismissing the claim, the court stated (pp 186-187): “The aim of defamation law is to protect individuals; a group may be sufficiently large that a státement concerning it cannot defame individual group members *** If the court were‘to permit an action to lie for the defamation of such a multitudinous group we would render meaningless the rights guaranteed by the First Amendment to explore issues of public import.” In Schuster v U.S. News & World Report (602 F2d 850) two well known distributors of Laetrile brought a libel action for damages resulting from *231the nationwide publication of two articles discussing the general controversy surrounding the drug, the illegal smuggling of it and the legal action resulting from the smuggling. Summary judgment for the defendants was affirmed. The United States Court of Appeals for the Eighth Circuit stated (p 854): “These vague and general references to a comparatively large group, without mentioning the plaintiffs, do not constitute an actionable defamation.” In Michigan United Conservation Clubs v CBS News (485 F Supp 893) the subject matter of a broadcast concerned the inhumane practices of hunters. Application of the comments was to a class of 1,000,000. The president of the Michigan Conservation Club sued, although neither he nor his club had been mentioned in the broadcast. The United States District Court stated that (p 900): “If plaintiffs were allowed to proceed with this claim, it could invite any number of vexatious lawsuits and seriously interfere with public discussion of issues, or groups, which are in the public eye. Statements about a religious, ethnic, or political group could invite thousands of lawsuits from disgruntled members of these groups claiming that the portrayal was inaccurate and thus libelous. Such suits would be especially damaging to the media, and could result in the public receiving less information about topics of general concern.”
In contrast to the treatment of an individual in a large group which has been defamed, an individual belonging to a small group may maintain an action for individual injury resulting from a defamatory comment about the group, by showing that he is a member of the group3 (see Neiman-Marcus v Lait, 13 FRD 311). Because the group is small and includes few individuals, reference to the individual plaintiff reasonably follows from the statement and the question of reference is left for the jury (Watts-Wagner Co. v General Motors Corp., 64 F Supp 506; see Note, Liability for Defamation of a Group, 34 Col L Rev 1322, 1325-1326).
*232In New York, the rules regarding reference to the individual in a context of group defamation were summarized by the Court of Appeals in Gross v Cantor (270 NY 93, 96, supra): “An action for defamation lies only in case the defendant has published the matter ‘of and concerning the plaintiff.’ (Corrigan v. Bobbs-Merrill Co., 228 N. Y. 58, 64.) Consequently an impersonal reproach of an indeterminate class is not actionable. (Eastwood v. Holmes, 1 F. & F. 347.) ‘But if the words may by any reasonable application, import a charge against several individuals, under some general description or general name, the plaintiff has the right to go on to trial, and it is for the jury to decide, whether the charge has the personal application averred by the plaintiff.’ (Ryckman v. Delavan, 25 Wend. 186, 202. So, Weston v. Commercial Advertiser Assn., 184 N. Y. 479; International Text-Book Co. v. Leader Printing Co., 189 Fed. Rep. 86; Wakley v. Healey, 7 C.B. 591. Cf. LeFanu v. Malcomson, 1 H. L. Cas. 637.)”
Gross (supra) concerned charges made by Eddie Cantor of dishonest artistic criticism against a group of 11 radio editors in New York City. Cantor moved to dismiss the complaint on the rationale that the criticism was directed at a class and had no personal application. The court refused, noting (270 NY, at p 96): “It does not there appear that the publication was so scattered a generality or described so large a class as such that no one could have been personally injured by it.”
Size per se is not stressed in Gross. The emphasis in the case and the New York authorities cited therein, is on distinguishing between a generalization applicable to a class and a specific charge directed at a circumscribed group of individuals. Accordingly, where a specific charge is made that the New York City Coroner’s office is graft ridden, the criminal acts described can be imputed to each of the four coroners and the four physicians in the office and each can maintain an action for defamaton. But where the charge is “in general terms that all the coroners in the state were a bad and corrupt lot”, it would not be defamatory “as against some individual who happened to be a member of the office somewhere” (Weston v Commercial Advertiser Assn., 184 NY 479, 485). The defamatory alie*233gation in the latter example fails to include sufficient particulars of identification to be actionable by an individual.
In Ryckman v Delavan (25 Wend 186, 199-202, supra), on which the Gross and Weston courts relied, and from which they quoted, the issue of individual application is discussed more fully: “Yet it does not thence follow, because a man is libelled — not by name or title or other specific description of himself, but under some such description of persons as includes certain other persons, and marks the individuality of each of them as much as if they were all severally named — that therefore this is no libel, having a personal application upon which a civil suit can be maintained. The application of the injurious charge to a particular person must be made on the same principle that the meaning of the charge itself is explained. Both are to be taken according to the common understanding of men and the customary use of language. * * * [I]f it appears clearly and undeniably on the plaintiff’s own declaratory statement of his case, that the charge was made against a whole body of men, such as lawyers, clergy or brewers, [etc.,] this is not a libel upon any individual of that class, and the courts must so pronounce it. But '^'L1--------1_------1 individuals, under some generj name, the plaintiff has the right to go on to trial, and it is for the jury to decide, whether the charge has the personal application averred by the plaintiff.” reasonable application, import
This analysis demonstrates that size is too narrow a focus to determine the issue of individual application in group defamation. Subsequent decisions in this area however have focused exclusively on size but no definitive limitation has been set. In Service Parking Corp. v Washington Times Co. (92 F2d 502, supra), the parking lot owners of downtown Washington were accused of using public streets to increase the rentable space of their lots. At the time 20 to 30 lots existed in the area and the lots were owned by 10 to 12 individuals. The court held that the size of the group was not so small as to cause the defamation of it to also defame the individual owners. In Neiman-Marcus v Lait (13 FRD 311, supra), a group of 25 salesmen was *234considered small enough to support individual actions to recover for a charge of homosexuality while a group consisting of 382 salesgirls was considered too large to permit individual actions to recover for allegations of prostitution. In the body of the decision, Service Parking Corp. v Washington Times Co. (supra) was cited as an example of the size of a group that was too large to support an individual action. No rationale was provided however to explain why an individual in a group of 25 was permitted to maintain an action for defamation in Neiman-Marcus while an individual in a group of 10 to 12 was not.
The argument of the media defendants in the instant case that the defamation of this particular group cannot support individual actions is based solely On the fact that the group exceeds 25. Their emphasis on a particular size finds some support in the commentary to section 564A of the Restatement of Torts, Second. Comment b provides: “It is not possible to set definite limits as to the size of the group or class, but the cases in which recovery has been allowed usually have involved numbers of 25 or fewer.” (See, also, Prosser, Law of Torts [4th ed], p 750; Sack, Libel, Slander, and Related Problems, §11.8.3.)
In New York there has been no articulated limit on size and we decline the opportunity to suggest one. An absolute limit on size is not justifiable. There is no compelling logic “in allowing a greater number of wrongs to afford a lesser degree of liability” in group defamation actions. (Lewis, The Individual Member’s Right to Recover for a Defamation Leveled at the Group, 17 Miami LQ 519, 535.) In Sumner v Buel (12 Johns 475), a case in which the majority opinion has since fallen into disrepute (1 Seelman, The Law of Libel and Slander in the State of New York, pars 98-100), the dissenting Justice Van Ness pointed out the inequitable and illogical results of applying the policy against a multiplicity of suits in actions of this type (p 482): “I cannot assent to the idea that the number of persons who may be libelled, affords the rule to determine whether or not an action will lie. Such a rule would be unjust and arbitrary. The libeller who calumniates a number of persons, by name, is liable to an action by each; and, in such a case, he would hardly be allowed to say, even in extenúa*235tion of his offence, much less in bar to the action, that, because he had exposed himself to so many actions, he ought not, therefore, to be punished at all. If such a rule should be adopted, the calumniator, who assails and reviles a great number of individuals in the same malicious publication, will escape; while the less guilty and less hardy slanderer, who has traduced the character of a single man only, shall be punished.”
Moreover, a limit on size does not necessarily protect frank discussions on matters of public concern. Church of Scientology of Cal. v Siegelman (481 F Supp 866) concerned a defamation action arising from the publication of a book on sudden and drastic alterations in personality in which the effects on personality caused by techniques used by religious cults and self-help therapies were studied. The action was allowed to proceed although the Scientology movement had over 5,000,000 members but the organizational instrumentalities or branch churches numbered 22.
Lastly, size limitations ignore the different characteristics of groups and the varied circumstances of publication which may permit relief for individual injury without interfering with First Amendment guarantees.
We are not the only court to find size limitations unduly restrictive. In Fawcett Pub. v Morris (377 P2d 42, 51 [Okla]) the Oklahoma Supreme Court could find “no substantial reason why size alone should be conclusive.” In Fawcett, the 60- to 70-man Oklahoma University football team was accused of taking amphetamines to improve performance. The team had had a victorious season and had won a postseason bowl game. The individual plaintiff was a fullback, a significant team position, and had played in all of the team’s games but two that season. He continued to play for two more years on the university team. He was also a member of the university baseball team.
The court held that the plaintiff had established his identity as one of the individuals libeled by the libel of the group, despite the size of the team, because the plaintiff was a constant player in a prominent position on the victorious team. In so doing the court rejected a definitive size limitation in favor of an intensity of suspicion test. *236(Note, Liability for Defamation of a Group, 34 Col L Rev 1322, 1324-1325; see, also, Developments in the Law — Defamation, 69 Harv L Rev 875, 894.)
The intensity of suspicion test recognizes “that even a general derogatory reference to a group does affect the reputation of every member” despite the size of the group (Note, Liability for Defamation of a Group, 34 Col L Rev 1322, 1324-1325). In order to determine personal application it requires that a factual inquiry be made to determine the degree that the defamatory comments or accusation of the group focuses on each of the individual members of the group. This test is not related to the exception to the rule against individual recovery for the defamation of large groups, which occurs where there is a special allusion to a particular member of the group or the circumstances surrounding publication give rise to the inference that there is a particular reference to a specific individual. (See National Nutritional Foods Assn. v Whelan, 492 F Supp 374; Fernicola v Farrar, Straus & Cudahy, 27 Misc 2d 565; Restatement, Torts 2d, § 564A, subd [b].) Under the exception, reference is to the individual under the guise of an attack on the group. Thus there is no need for an intensity of suspicion test in these circumstances.
With the intensity of suspicion test, size is a consideration and the probability of recovery diminishes with increasing size. Size however, is not the only factor evaluated. It is balanced against the definiteness in number and composition of the group and its degree of organization (Note, Liability for Defamation of a Group, 34 Col L Rev 1322, 1326). This list of balancing factors or reference elements is not meant to be exclusive. The variations in the types of groups is too great (Note, Liability for Defamation of a Group, 34 Col L Rev 1322, 1325). Thus the case of Fawcett Pub. v Morris (377 P2d 42 [Okla], supra) correctly focused on reference elements not articulated prior to its analysis. Those reference elements were the prominence of the group and the prominence of the individual within the group.
On these cross motions for summary judgment, we must decide whether the plaintiff can maintain an action given the size and characteristics of the group defamed. To *237achieve that end, we utilize the intensity of suspicion test. The test is consistent with New York law. It provides a flexible approach for balancing the competing interests present in group defamation actions. The proper application of the reference elements of the test will prevent vexious actions based on impersonal discussions of matters of public concern without terminating suits on an arbitrary size limitation where real injury is present.
We adopt the reference elements demonstrating personal application that have been articulated prior to this decision. We have developed the relationship between the elements and have added refinement to the underlying concept.
Initially, the group to which the allegedly defamatory comment refers must be isolated by the standards set forth or implied in the comment. The plaintiff must be within that group. Imputation to the plaintiff will be evaluated in relation to the group as defined by the comment and not by the plaintiff’s relationship to a smaller subset of the group defined (United Med. Labs. v Columbia Broadcasting System, 404 F2d 706; Mario’s Enterprises v Morton-Norwich Prods., 487 F Supp 1308; Loeb v Globe Newspaper Co., 489 F Supp 481).
Here the group is explicitly defined as the unindicted police officers of the City of Newburgh in 1972. The plaintiff is a member of this group and is still employed as a police officer. The limitations of time, place, employment and lack of criminal status are evidence that a group of individuals was mentioned under a common description rather than a class. Incidentally, it should be noted that the explicit reference in the comments to the group prevents a libel on government challenge.4
Size is the first reference element to be examined. In this case the group consists of at least 53 individuals. The size *238of the group as compared to the 600,000,000 Muslims in Mansour v Fanning (506 F Supp 186, supra) is obviously not so large that it is unmeasurable. While the number of policemen is larger than the group of 25 salesmen whose members were permitted to maintain actions in Neiman-Marcus v Lait (13 FRD 311, supra) it is also more prominent and it is substantially smaller than the group of 382 saleswomen who were not. In our opinion although the difference between at least 53 individuals and 25 is measurable, it is not significant in terms of the public perception of the size of the group. Public perception of the size of the group is an important factor in focusing on the degree of suspicion attributed to individual members of the group irrespective- of the actual size of the group. Unless the recipient of the defamatory allegations can perceive the size of the group to which the allegations refer, the recipient cannot judge whether to apply the observation that the comment is less likely to be true with respect to every member of a large group. Moreover, given the nature and specificity of the charge herein and its compelling logic, the increased size is not sufficient to dilute the harm caused to the individuals from the statement so that the resulting injury falls below the threshold of legal recognition.
The group of at least 53 unindicted police officers enjoys a high degree of organization. This group is not dependent on the editorial for the existence of its group concept unlike the parking lot owners in Service Parking Corp. v Washington Times Co. (92 F2d 502, supra). Nor have the comments herein imposed a class organization on otherwise scattered and unrelated individuals as occurred with the Laetrile distributors in Schuster v U.S. News & World Report (602 F2d 850, supra). The unindicted police officers are related by employment in a particular city at a particular time and enjoy an existence as a group entity independent of the accusations of criminal misconduct. At the time of the scandal, this group represented the entire police force of Newburgh. A similar set of circumstances arose in Fawcett Pub. v Morris (377 P2d 42 [Okla], supra) where the defamatory comments applied to the existing football team of a particular university and an individual player was *239permitted to maintain an action for harm resulting from defamation of the team.
The composition of the group in the instant case was definite and its size was fixed in 1972. Entry into the group was restricted and membership was highly visible by virtue of badges and Uniforms during a period of intense scrutiny caused by the scandal. These qualities of visibility and lack of easy access are similar to those demonstrated by the radio critics in Gross v Cantor (270 NY 93, supra) where an individual was permitted to set before the jury the question of individual reference.
Since 1972, however, the group defamed has not been coextensive with the police force of Newburgh. Replacements for the indicted officers have undoubtedly been made although there is no indication in the record of a constantly changing roster as there was in Fawcett (supra). Thus the visibility of the defamed group within the police force has diminished over time. With decreased visibility comes the possibility that the comments will be mistakenly applied to replacements and cause injury to their reputations and effectiveness as police officers. This possibility that the defamatory comments will be applied to the present police force as a whole, does not interfere with a finding that the comments are capable of personal application to the individual members of the defamed group, because over the years, the membership of the defamed group has remained constant and is easily discernible. With renewed attention focused on the group entity as it existed in 1972, as a result of the editorial, the membership of the defamed group has once again become apparent.
The last reference element that must be analyzed is the prominence of the group and its individual members. Prominence of the group must be considered generally within the geographic area of publication or distribution of the defamatory allegation (Lewis, The Individual Member’s Right to Recover for a Defamation Levelled at the Group, 17 Miami LQ 519, 527) and in the context of the allegation’s subject matter. Applying these concepts, we note that police officers in.small communities are gener*240ally prominent public officials5 and Newburgh is a small city “the kind of [place] where people know each other” (Hinsdale v Orange County Pub., 17 NY2d 284, 290).
As we have noted the prominence of this group was intensified by public scrutiny at the time of the scandal throughout the distribution area of the newspaper. During the scandal, the newspaper’s editorials twice focused public attention on the unindicted remaining police force, each time calling for additional indictments because the unindicted members were aware of the misconduct in the rest of the department. The last editorial, which is the subject of this suit, published seven years after the first announcement of the misbehavior, refocused public attention, incorporating all the prior publicity and renewing the prominence of the unindicted members of Newburgh’s police force. Moreover, this compulsive repetition of the charge by the newspaper gives rise to the inference that the newspaper had access to special information which would negate the possibility that the charge of misconduct was an inexact generalization (Developments in Law — Defamation, 69 Harv L Rev 875, 895).
With respect to the prominence of the individuals within the group, we are in a different position than the court in Fawcett Pub. v Morris (377 P2d 42 [Okla], supra) which was decided on a full trial record. We have no evidence of the personal prominence of the plaintiff police officer, except for his long-term employment, which for the purposes of these cross motions is sufficient. Our inquiry is different than the Fawcett court. Here we decide the minimal factual development of the issues, whether personal application of the comments is possible as a matter of law, while the Fawcett court decided on the trial evidence whether acttial reference had been shown. We conclude *241that the editorial comments are capable of personal application.
If there is a renewal of the cross motions, evidence relevant to the factors we have isolated should be presented, as well as evidence by plaintiff of personal application.
THE NEW OPINION PRIVILEGE
At Special Term, the media defendants argued that the allegedly defamatory statements were fair comments based on facts stated, and therefore privileged. However, there are elements of the privilege not articulated in the argument. A comment is fair when it is based “‘on facts truly stated and free from imputations of corrupt or dishonorable motives on the part of the person whose conduct is criticized, and is an honest expression of the writer’s real opinion or belief.’” (Julian v American Business Consultants, 2 NY2d 1, 8, supra.)
Without doubt, there is no mode of argument which would free the defamatory statements herein from the imputations of corrupt or dishonorable conduct contained in them. Thus the media defendants have abandoned the fair comment argument in this court.
On appeal, they argue that the statements are opinions and are privileged by the new opinion privilege. The argument is based on a passage in Gertz v Robert Welch, Inc. (418 US 323, 339-340): “We begin with the common ground. Under the First Amendment there is no such thing as a false idea. However pernicious an opinion may seem, we depend for its correction not on the conscience of judges and juries but on the competition of other ideas.” As the United States Court of Appeals for the Second Circuit has noted, this passage “has become the opening salvo in all arguments for protection from defamation actions on the ground of opinion, even though the case did not remotely concern the question” (Cianci v New Times Pub. Co., 639 F2d 54, 61). The Court of Appeals, in Rinaldi v Holt, Rinehart & Winston (42 NY2d 369) was aware of this passage. Nevertheless, with respect to allegations of corruption, it unequivocally stated (p 382) that “[accusations of criminal activity, even in the form of opinion, are not *242constitutionally protected.” The rationale underlying this decision is that accusations of criminal behavior are factually laden opinions whose truth or falsity can be tested (Cianci v New Times Pub. Co., 639 F2d 54, supra). Thus they are treated as statements of fact, protected only by the privilege generated by adherence to the constitutionally mandated standard of care (Rinaldi v Holt, Rinehart & Winston, supra, p 382). Therefore, no opinion privilege can be extended to protect the editor’s accusations in this case, even assuming that such a privilege exists and replaces the privilege of fair comment.
CONCLUSION
Without the protection of either privilege, the issue of the truth or falsity of the comments must be determined. Since the editorial involves a topic of public concern, the plaintiff has the burden of showing that the comments were false (Fairley v Peekskill Star Corp., 83 AD2d 294). His sworn statement that he did not know of the misconduct in his department is insufficient to prove that the comments were false as they relate to him (Rinaldi v Holt, Rinehart & Winston, supra, p 382). Moreover, if the defamatory statement accusing the entire group of knowledge of the misconduct in the department is substantially true, the falsity of the statement as it relates to an individual policeman may be irrelevant (Note, Liability for Defamation of a Group, 34 Col L Rev 1322, 1327-1329). Here however, the policeman’s deficiency in proof of falsity is not fatal to his action. The media defendants’ position on these cross motions was that the statements were opinion and that, therefore, they could not be true or false.
The question of actual malice must also await further discovery. Like falsity, the plaintiff has the burden on this question when it is in issue (Rinaldi v Holt, Rinehart & Winston, supra, p 384). Here, however, the question of whether the comments were published with knowledge of falsity or in reckless disregard of the truth was not ripe since, as noted above, the media defendants argued that the comments were opinions whose truth or falsity was irrelevant.
In conclusion, the cross motions for summary judgment were properly denied. The parties may renew, however, if *243they choose, upon completion of discovery. It is necessary to stress the limits of our holding. We find merely that the plaintiff is not barred from litigating because he is a member of a group exceeding 25 individuals and that the defendants are not protected from liability by a constitutional exception for statements of opinion. This holding in no way relieves the plaintiff of his substantial burden on the issues of falsity and actual malice on the renewal of the cross motions.
Mollen, P. J., Hopkins and Weinstein, JJ., concur.
Order of the Supreme Court, Orange County, dated August 13, 1980, affirmed insofar as appealed from, without costs or disbursements, without prejudice to renewal of the cross motions for summary judgment when discovery has been completed.
. The defendants assert, without explanation but also without “numerical” challenge, that the total number of police officers at that time was over 70. If 18 were indicted, the remaining force numbered at least 53.
. The psychological assumption underlying the rule is not without challenge. Where a reader irrationally holds certain prejudices against the group defamed, the defamatory comments reinforce those prejudices irrespective of size. Thus the target group and its individual members suffer injury (Group Vilification Reconsidered, 89 Yale LJ 308, 311-313).
. For the purposes of this discussion, it is not necessary to analyze the law applicable to situations where only a portion of the group is defamed. (See Arcand v Evening Call Pub. Co., 567 F2d 1163; Prosser, Law of Torts [4th ed], p 751; Neiman-Marcus v Lait, 13 FRD 311.)
. Group defamation often arises in the context of a criticism of government. Permitting recovery for an impersonal reproach of governmental conduct would constitute an unconstitutional libel on government. Thus explicit reference to a specifically defined group is necessary before the comments can serve as the basis of a defamation action by those performing or administering the functions criticized (Rosenblatt v Baer, 383 US 75; New York Times Co. v Sullivan, 376 US 254). There is no constitutional impediment of this nature here. As noted, the defamatory statements refer explicitly to a specific group of police officers, although they were published in an impersonal criticism of proposed legislation by the City Council.
. The plaintiff police officer asserts that he is not a public official for the purpose of a defamation action. The matter is well settled in this State, a police officer is a public official (Orr v Lynch, 60 AD2d 949, affd 45 NY2d 903; Malerba v Newsday, 64 AD2d 623). While the average police officer does not have “substantial responsibility for or control over the conduct of governmental affairs”, he is in a position which “has such apparent importance that the public has an independent interest in the qualifications and performance of the person who holds it, beyond the general public interest in the qualifications and performance of all government employees” (Rosenblatt v Baer, 383 US 75, 85-86).