Today, a majority of this court, upon reargument, adheres to its reversal in Matter of Katz v. Hoberman and Matter of Elliott v. Haberman (28 N Y 2d 530). I dissented from the original determination for the reasons expressed in the opinion at Special Term (61 Misc 2d 411) and, now again dissent on the opinion at Special Term. I add only the following comments.
There is no factual dispute in this case. Respondents, lieutenants in the New York City Police Department, New York City Transit Police Department and New York City Housing Authority Police, are civil service employees and were eligible to take promotional examinations to the position of captain in their respective departments. On January 18, 1969 appellants conducted such an examination at which time respondents competed. On March 18, 1969 before any of the papers had been graded, the City Department of Personnel issued a press release in which the January examination was canceled and a new examination was scheduled for May 24, 1969. It was stated therein, *973that 12 questions on the January examination were originally taken from a 1959 California police promotional examination which the examiner had thought was unpublished but that an investigation revealed that' the questions were identical to those published in the 1962 Police Yearbook, and since that publication had been available in the New York area for several years, it was concluded by the commission that: “ 1 This destroys the competitive nature of the examination, raises questions concerning the integrity of the examining process and reduces reliability of the test, all in violation of the New York State Constitution. ’ ”
Upon the cancellation of the January examination and the scheduling of a new examination, respondents commenced the instant article 78 proceedings.
The Appellate Division (33 A D 2d 753), by affirming orders of Special Term, found that appellants had acted arbitrarily because the use of the 12 questions did not impair the competitiveness of the examination (61 Misc 2d 411). A majority in this court reversed, and reinstated the determination of the Civil Service Commission (28 N Y 2d 530). It reasoned that appellant’s action in canceling the examination could not be considered arbitrary because, vested with broad administrative discretion, the commission could reasonably determine that the competitiveness of the examination was impaired by the possibility that some candidates may have used the 1962 Police Yearbook in their preparation for the examination. The court has placed reliance on our decision in Matter of Chironna v. Watson (304 N. Y. 255, mot. for rearg. den. 304 N. Y. 738). This case is, however, simply not in point. There, after an open book examination had been administered and graded, it developed that five questions in that examination had been printed in a reference book, and that some candidates might have had access to this book in the examination room. It was conceded there, however, that this factor rendered the examination noncompetitive; hence, the sole question before the court was not that of competitiveness but the propriety of the corrective measures employed by the Civil Service Commission. Unlike Chironna, the examinations involved in this case were not “ open book,” and we are required to pass on the basic issue of competitiveness.
Now, on reargument, the majority attempts clarification by adding that this decision is to have no adverse effect upon the accepted practice of utilizing previous examination questions *974in the preparation of subsequent examinations. Further, it finds some significance in the fact that the disputed 12 questions were grouped together, rather than interspersed throughout the examination. If prior publication of questions is to be the criteria for a declaration of noncompetitiveness, then it should make no difference where they appear on the examination. Hence, I remain unpersuaded by this hypertechnical rationale employed by the majority.
In reviewing the administrative action taken against respondents, the basic question presented is, of course, whether the cancellation of the examination was arbitrary, capricious or unreasonable. Our Constitution provides that civil service appointments and promotions must ‘ ‘ be made according to merit and fitness to be ascertained, as far as practicable, by examination which, as far as practicable, shall be competitive” (N. Y. Const., art. V, § 6; see Civil Service Law, § 50; Matter of Fink v. Finegan, 270 N. Y. 356). Appellants have been charged with the performance with this mandate. Although they are vested wide latitude of discretion in this regard (see Matter of Wirzberger v. Watson, 305 N. Y. 507, 513), it is by no means unfettered and is subject to review by this court. The reason for this is clear. Inasmuch as the competitiveness of civil service examinations is a concept embodied in our Constitution, decisions, such as those made by the appellants herein, are to be treated as questions of law (see CPLR 7803, subd. 3). Stated differently, appellants ’ discretion in these matters does not give them the power to act without standards and it is the Constitution, as construed by the courts, which is the source of the criteria which are to be employed by civil service commissions in determining whether or not a particular examination is defective.
Thus, if the examinations involved in the cases at bar remained competitive, despite the prior publication, then appellants have abused their power as a matter of law since their action is devoid of a viable standard. This question was answered in the affirmative by the courts below; and, as I have indicated, I am in complete agreement with the conclusions that no proof was adduced which would establish such infirmity in the examination.
It is significant here that neither the Civil Service Commission nor for that matter any of the other appellants, had any proof that any candidate for promotion had seen the Police *975Yearbook before the examination was administered. Indeed, appellants’ action seems predicated on the mere possibility that this was in fact what occurred. I am unable to see how this speculative assumption can be viewed as being sufficient to sustain their action. But even assuming that all candidates had read and studied the Police Yearbook, the competitiveness of the examination would nevertheless remain unimpaired. As is abundantly clear, prior to the examination candidates were given a suggested bibliography and specifically advised in a circular prepared by the New York City Police Academy and acquiesced in by the commission, that: ‘£ The forthcoming written tests for promotion in the police service (NYCPD, Transit PD, Housing PD) will be of the multiple-choice type. The questions in* these tests will be substantially based on, but not restricted to, materials contained in the authoritative texts, periodicals and publications listed here. Note the fact that although these many sources are listed, candidates should not assume that equal emphasis will be given to each source, nor that the order of the sources is related to the order of the questions on the test, nor that all the listed materials will in fact be used, nor that candidates are expected to be conversant with every word in every text or reference listed. This list has been prepared solely for your convenience and as a guide to your study efforts.” (Emphasis supplied.)
It is my view that upon receipt of this information, a prudent candidate for promotion would immerse himself in the materials listed and other relevant sources so as to prepare for the examination. Though the Police Yearbook was not named in the bibliography, there is no question but that it qualifies as an appropriate study material. Inasmuch as good preparation often involves the examination and study of the type of question which is usually employed by the examiners, I am at a loss to understand how efficient preparation for an examination would render it noncompetitive.
All candidates had this opportunity and I consider it manifestly arbitrary and unfair to penalize those whose study and research involved months of effort, particularly where, as here, the basic problem was created by the appellants and not by any candidate. In every examination, there are always those who are more diligent in their preparation. This is why some are successful and others are not. Some candidates, in the hope of *976attaining a passing grade, enroll in courses which are specifically geared to the study of materials upon which the examination is based. The majority’s decision in these cases would seem to suggest that if an instructor should happen to mention questions which, fortuitously, are later found on the examination, appellants would be justified in canceling the examination as noncompetitive. This is untenable and jeopardizes other types of examinations. For example, candidates for admission to the Bar of this State have, for years, studied questions which purportedly were on old examinations. The decision reached by the court would appear to cast serious doubt on this long-accepted practice. Though the court sanctions the use of prior examination questions, it nevertheless concludes that if it is later discovered that they have legitimately entered the public domain, the examination is no longer competitive. Consequently, if a candidate wishes to refer to a sample-type question as a part of his preparation he must do so at the risk of a subsequent determination that the examination has become noncompetitive for that reason. This not only inhibits effective preparation, but also necessarily entails a senseless duplication of effort, on the part of both the candidate and the examiners. These considerations should not be overlooked in reaching a just determination of the cases at bar.
Accordingly, I am unable to agree that the prior publication of the 12 questions rendered the examination noncompetitive and conclude that the courts below have correctly determined that the cancellation of the examination was arbitrary, capricious and an abuse of discretion.
Chief Judge Ftjld and Judges Burke, Bergak, Breitel, Jasek and Gibson concur in memorandum; Judge Scileppi dissents and votes to affirm the order appealed from in an opinion.
In each case: Upon reargument, original decision of reversal dated January 7, 1971 adhered to in a memorandum.