[1993] 1 F.C. 427
T-1418-92
The Information Commissioner of Canada (Applicant)
v.
The Prime Minister of Canada (Respondent)
T-1867-92
The Information Commissioner of Canada (Applicant)
v.
The Prime Minister of Canada (Respondent)
T-1524-92
Mary Calamai (Applicant)
v.
The Prime Minister of Canada and the Privy Council Office (Respondents)
T-1390-92
Ken Rubin (Applicant)
v.
The Clerk of the Privy Council (Respondent)
Indexed as: Canada (Information Commissioner) v. Canada (Prime Minister) (T.D.)
Trial Division, Rothstein J.—Ottawa, August 11 and November 19, 1992.
Access to information — Access sought to Government-commissioned public information polls on national unity and constitutional reform — Application to review Government's refusal to disclose most of information sought — Threshold test: “reasonable expectation of probable harm”, not “any potential for injury” — No “sliding scale” according to importance of information involved — First case where Access to Information Act, s. 14 exemption (disclosure injurious to conduct of federal-provincial affairs) invoked — Role of Court under Act, s. 50 — Presumption in favour of disclosure — Evidentiary requirement — No reasonable grounds to withhold disclosure of record based on evidence presented by Government.
Constitutional law — Charter of Rights — Fundamental freedoms — Freedom of press — Access sought to Government-commissioned public information polls on national unity and constitutional reform — Whether Access to Information Act, s. 14 should be interpreted in light of Charter, s. 2(b) — Not mere matter of statutory interpretation — If s. 14 expressly confers power on government to limit assumed protected right (access to government information), attack must be on constitutional validity of statutory provision — Charter challenge not dealt with as Federal Court Act, s. 57 requirement of notices to attorneys general not complied with.
Requests for information sought records of Government-commissioned public information polls and results of focus or survey groups and relevant contracts concerning national unity and constitutional reform for the period July 1, 1990 to December 31, 1991. Of 709 pages identified by the Access to Information/Privacy Coordinator for the Privy Council to be relevant, only 120 pages were disclosed. The Information Commissioner found that the complaints were well founded and that all but 74 pages should be released. This was an application to review the refusal by the Government to disclose the requested record. The refusal was based on section 14 of the Access to Information Act which provides that disclosure may be denied if injurious to Government strategy with respect to the conduct of federal-provincial affairs. This was the first time that the Court had been called upon to consider that exemption. The Government argued that the stakes were so high (the very existence of the country) that nothing could justify, at this time, releasing information that could a) constrain the Government’s strategic and tactical options, b) have a potentially negative impact on the environment for the federal-provincial negotiations, c) prejudice the ongoing discussions with the provinces and, d) furnish information to those hostile to Canadian unity.
Held, the applications should be granted with costs.
Subsection 2(1) of the Access to Information Act establishes guiding principles for the interpretation of the operative provisions of the Act: a) government information should be available to the public; b) necessary exemptions should be limited and specific; c) decisions on disclosure should be reviewed independent of government.
Section 14 of the Act, under which the exception was claimed, uses the words “could reasonably be expected to”. In Canada Packers Inc. v. Canada (Minister of Agriculture), the Federal Court of Appeal interpreted those words, as used in paragraph 20(1)(c), as meaning that the exception to access must be based on a “reasonable expectation of probable harm”. This Court was bound by that decision. The more pro-government Australian approach, with its sliding scale test and more literal statutory interpretation, could not be accepted, given the clear and direct Canadian authority in this area of the law. The Canadian case law interpreting the Access to Information Act has established guidelines that can be useful in assessing whether or not there is a reasonable expectation of probable harm from disclosure in a given situation and the procedures to be followed.
Arguments based on paragraph 2(b) of the Charter (freedom of expression) were considered as an attack on the constitutional validity of section 14 of the Access to Information Act (assuming, without deciding, that paragraph 2(b) applied to the type of government information under consideration). It was not merely a matter of statutory interpretation (interpreting section 14 in light of paragraph 2(b) of the Charter). Applying the Supreme Court of Canada decision in Slaight Communications Inc. v. Davidson, it was found that a decision under section 14 was confined to the formulation of an opinion as to whether or not disclosure could reasonably be expected to be injurious, and involved little if any discretion. Therefore, the Government’s power to limit an assumed protected right (access to government information) would be conferred by section 14 and the attack must be on the constitutionality of that section. Since the notice required by section 57 of the Federal Court Act was not given, the Court refused to adjudicate that Charter challenge.
The Federal Court’s process under section 50 of the Access to Information Act relies on evidence. It is not an expert tribunal with its own resources and must act on the basis of the evidence before it. Therefore, the heavy onus placed on the party seeking to maintain confidentiality must be satisfied in a formal manner on a balance of probabilities through clear and direct evidence.
With the enactment of the Access to Information Act in 1982 and the repeal of subsection 41(2) of the Federal Court Act (which gave the Government the absolute right to maintain confidentiality in federal-provincial relations matters), Parliament demonstrated that it was no longer satisfied with an unjustifiably cautious approach by Government to disclosure. The question herein was whether the Government had reasonable grounds to expect harm from disclosure or whether it was taking an overly cautious approach to disclosure, based on possibility, which was not justified by the legislation.
It could be inferred from subsection 2(1) of the Act that a general approach to justifying confidentiality was not envisaged. There must be a clear and direct linkage between the disclosure of specific information and the harm alleged. And it must be explained how or why the harm alleged would result from disclosure of specific pages of the record.
On the basis of the evidence of the Government witnesses, it was concluded that: (1) Their approach indicated a well-intentioned attempt to avoid risk rather than to assess a reasonable expectation of harm from disclosure. (2) There was little in the way of specific evidence linking a reasonable expectation of harm to the contents of specific pages. (3) There was only general reference or no express reference by the Government witnesses to approximately 90% of the record. This was insufficient to enable the Court to conclude that there were reasonable grounds for confidentiality in respect of this information. (4) There was no specific assessment of the numerous overlaps between public and Government information.
STATUTES AND REGULATIONS JUDICIALLY CONSIDERED
Access to Information Act, R.S.C., 1985, c. A-1, ss. 2, 4, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 41, 42, 47, 48, 50.
Access to Information Act, S.C. 1980-81-82-83, c. 111, Sch. 1.
Canada Labour Code, R.S.C. 1970, c. L-1.
Canadian Charter of Rights and Freedoms, being Part I of the Constitution Act, 1982, Schedule B, Canada Act 1982, 1982, c. 11 (U.K.) [R.S.C., 1985, Appendix II, No. 44], s. 2(b).
Federal Court Act, R.S.C. 1970 (2nd Supp.), c. 10, s. 41(2) (rep. by S.C. 1980-81-82-83, c. 111, s. 3).
Federal Court Act, R.S.C., 1985, c. F-7, s. 57 (as am. by S.C. 1990, c. 8, s. 19).
Freedom of Information Act, No. 3, 1982 (Aust.), s. 3, 33A(1) (as enacted by Freedom of Information Amendment Act, 1983, No. 81, s. 17).
CASES JUDICIALLY CONSIDERED
FOLLOWED:
Canada Packers Inc. v. Canada (Minister of Agriculture), [1989] 1 F.C. 47; (1988), 53 D.L.R. (4th) 246; 32 Admin. L.R. 178; 26 C.P.R. (3d) 407; 87 N.R. 8 (C.A.); Saint John Shipbuilding Ltd. v. Canada (Minister of Supply and Services) (1990), 67 D.L.R. (4th) 315; 107 N.R. 315 (F.C.A.).
APPLIED:
Maislin Industries Limited v. Minister for Industry, Trade and Commerce, [1984] 1 F.C. 939; (1984), 10 D.L.R. (4th) 417; 8 Admin. L.R. 305; 27 B.L.R. 84 (T.D.); Slaight Communications Inc. v. Davidson, [1989] 1 S.C.R. 1038; (1989), 59 D.L.R. (4th) 416; 26 C.C.E.L. 85; 89 CLLC 14,031; 93 N.R. 183.
DISTINGUISHED:
Arnold (on behalf of Australians for Animals) v Queensland (1987), 73 ALR 607 (Aust. Fed. Ct.).
REFFERED TO:
Rubin v. Canada (Canada Mortgage and Housing Corp.), [1989] 1 F.C. 265; (1988), 52 D.L.R. (4th) 671; 19 F.T.R. 160; 86 N.R. 186 (C.A.); Air Atonabee Ltd. v. Canada (Minister of Transport) (1989), 27 C.P.R. (3d) 180; 27 F.T.R. 194 (F.C.T.D.); Ottawa Football Club v. Canada (Minister of Fitness and Amateur Sports), [1989] 2 F.C. 480; (1989), 23 C.P.R. (3d) 297; 24 F.T.R. 62 (T.D.) ; Montana Band of Indians v. Can. (Min. of Indian & Nor. Affairs), [1988] 5 W.W.R. 151; (1988), 59 Alta. L.R. (2d) 353; 18 F.T.R. 15 (F.C.T.D.); Ternette v. Canada (Solicitor General), [1992] 2 F.C. 75; (1991), 86 D.L.R. (4th) 281; 39 C.P.R. (3d) 371; 49 F.T.R. 161 (T.D.); Merck Frosst Canada Inc. v. Canada (Department of Health and Welfare Protection Branch) (1988), 20 C.P.R. (3d) 177 (F.C.T.D.).
AUTHORS CITED
Johnston, Richard et al. Letting the People Decide: Dynamics of a Canadian Election, Montréal: McGill-Queen’s University Press, 1992.
APPLICATIONS to review the refusal of the Government to disclose certain records (Government-commissioned public opinion polls on national unity and constitutional reform between July 1, 1990 and December 31, 1991) on the ground that disclosure of the information could be injurious to the conduct of federal-provincial affairs (section 14 of the Access to Information Act). Applications granted.
COUNSEL:
David W. Scott, Q.C., Peter K. Doody and Daniel Brunet for applicant Information Commissioner.
Richard G. Dearden and Neil Wilson for applicant Mary Calamai.
Barbara A. McIsaac and T. Bradbrooke Smith for respondents.
APPAREANCE:
Ken Rubin on his own behalf.
SOLICITORS:
Scott & Aylen, Ottawa, for applicant Information Commissioner.
Gowling, Strathy & Henderson for applicant Mary Calamai.
Deputy Attorney General of Canada for respondents.
APPLICANT ON HIS OWN BEHALF:
Ken Rubin.
The following are the reasons for order rendered in English by
Rothstein J.:
A. INTRODUCTION
This case arises out of four applications brought under the Access to Information Act, R.S.C., 1985, c. A-1 (the Act):
1. Application dated June 15, 1992 by the Information Commissioner of Canada (herein referred to as the Information Commissioner or Commissioner) against the Prime Minister of Canada[1] (Court No. T-1418-92) for a review of the refusal of Government to disclose certain records requested by Jeffrey Simpson on November 19, 1991 being:
All records on public opinion polls and focus groups on constitutional matters from July 1, 1990 to September 24, 1991.
and by Kirk Lapointe on November 19, 1991 being:
All records on public opinion polls and focus groups on constitutional matters from September 25, 1991 to the present [to November 19, 1991].
2. Application dated July 22, 1992 by the Information Commissioner against the Prime Minister (Court No. T-1867-92) for a review of the refusal of the Government to disclose certain records requested by Jason Morris on January 30, 1992 being:
In P.C.O. file 929178 “National Unity Polls” regarding constitutional changes from January 1, 1991 to December 30, 1991.
and by Norman Nielsen on January 28, 1992 being:
In P.C.O. file 929179 “Results of public opinion polling” conducted in 90/91 on national unity issues and the constitution.
3. Application dated June 23, 1992 by Mary Calamai against the Prime Minister and the Privy Council Office (herein referred to as the PCO) (Court No. T-1524-92) for a review of the Government’s refusal to allow Ms. Calamai access to information she requested on November 7, 1991 being:
I request access to the results of all polls, surveys, focus groups or other forms of public opinion sampling on the subject of the constitution or constitutional reform carried out since July 1, 1990....
4. Application filed June 15, 1992 by Ken Rubin against the Clerk of the Privy Council (Court No. T-1390-92) for a review of the refusal of the Government to disclose to Mr. Rubin information he requested on November 29, 1991 being:
A descriptive list of all polls since January 1991 commissioned by the PCO, their costs, contracts, and copies of the interim/final results or any records about these polls. Dec/91 or when this response is provided.
Although each request for information was different in detail, in total they sought records of Government-commissioned public opinion polls and results of focus or survey groups and relevant contracts all related to the subject of national unity and constitutional reform over the period July 1, 1990 to December 31, 1991.
The way in which each request for information was handled was essentially the same. The Access to Information/Privacy Coordinator for the Privy Council identified 709 pages as being relevant to the access request. Of these, 120[2] pages were disclosed and provided to the applicants. Disclosure of some 589 pages[3] was refused. In each case the requesters complained to the Information Commissioner about the refusal.
On January 16, 1992 the Commissioner began an exchange of correspondence with the Clerk of the Privy Council. All relevant records and information were provided by the PCO to the Information Commissioner. The Information Commissioner received representations from the requesters and from the Government, notified the parties opposite in interest of the positions being taken by the other and offered all parties further opportunities to make submissions.
On April 15, 1992 the Information Commissioner reported to the Clerk of the Privy Council that he found that the complaints were well-founded and further that all requested material, with the exception of 74 pages, should be released. The Commissioner requested the Clerk of the Privy Council to inform him by May 11, 1992 as to whether or not the Commissioner’s recommendations would be implemented and if so to what extent.
On May 13, 1992, the Clerk of the Privy Council responded to the Commissioner and provided reasons why he did not propose to take action on the Commissioner’s recommendations.
In June 1992, the Information Commissioner informed the requesters of his findings, conclusions and recommendations, and his willingness, with the consent of the requesters, to apply to this Court for a review of the refusal by the Government to disclose the requested records.
Simpson and Lapointe and later Morris and Nielsen authorized the Information Commissioner to apply to the Federal Court. Ms. Calamai and Mr. Rubin did not do so because they considered that all the information that had been requested (and not just that information sought by the Information Commissioner) should be released.
The reasons for all four orders are given in Canada (Information Commissioner) v. Canada (Prime Minister), Court No. T-1418-92. They take into account the evidence and arguments made in respect of each of the four motions.
B. FINDINGS OF THE INFORMATION COMMISSIONER
The Information Commissioner’s findings of April 15, 1992 outline the difficulty in this case:
There is no doubt that, in the current volatile context of discussions on constitutional renewal, ample justification exists for a cautious approach to information disclosure. A misspoken, misinterpreted or mistimed phrase may cause difficulties for ongoing federal-provincial negotiations on unity. Frankly, it is enormously discomforting for me to have to deal with the question these complaints raise, namely, has PCO been so cautious about disclosure as to fall outside the bounds of the “reasonable expectation of injury” test contained in section 14. Alas, the law thrust that task upon me.
The Information Commissioner then outlined the difficulty he perceived in this case. On this point the Information Commissioner states:
On one point, I have no doubt. Even in the context of our troubling constitutional ills, where the stakes (the very existence of the country as we have known it) are so high, the threshold test to justify secrecy has not changed. While some might wish to withhold records which have “any potential” for injury if disclosed, the law simply does not, in my view, offer the government (or me) such latitude.
That being said, I also have no doubt that the government’s decision to withhold the records at issue in these complaints was taken in good faith, after careful review at the highest levels and in the context of a general commitment to openness ....
The Information Commissioner’s findings and recommendations follow:
Having personally reviewed the withheld records under investigation here and considered the arguments for and against their disclosure, I am not able to be as doctrinaire as I have previously been. In these cases, I am satisfied that some portions of the records, if disclosed, could reasonably be expected to result in the injury described in section 14.
Portions could give insight into strategic alternatives being considered by the government and even an Information Commissioner can see the legitimate need for the government to keep some of its cards close to the vest in sensitive negotiations. And, yes, there are lines of inquiry in the records—designed to probe the depth and limits of the public’s reaction to certain ideas, individuals, groups and scenarios—which by their nature may be inelegantly or clumsily phrased, at least from the perspective of public consumption. It is not unreasonable, in my view, to accept that, if disclosed, such information could disrupt the already troubled atmosphere in which the constitutional renewal exercise is being played out.
Having recognized all that, I must report that my personal review of the survey records has also convinced me that, if some information certainly qualifies for exemption, much does not. For the most part, the questions, answers and analyses deal with public issues and public proposals. I’m simply not able to agree that disclosure of such information could compromise the government’s strategic or tactical options.
...
To return to the formal legal position I must tell you that, in my view, the majority of the withheld records could not reasonably be expected to injure the conduct of federal-provincial relations if disclosed. In particular, I believe there should be disclosure of questions and answers about matters publicly known to be under consideration in the unity discussions or which reveal the survey responses to questions about publicly announced policies, proposals, strategies or ideas (be they federal, provincial or from other groups or individuals).
True, if such information were disclosed, it might well become the subject of media or other public comment and it could be used to support or criticize the positions of various players in the unity debate. To such a concern I would answer with the government’s own view that a significant level of informed public involvement in the debate is necessary to avoid the perception of public exclusion—a perception, which, I would argue, itself carries a heavy risk of injury to federal-provincial relations. And one must bear in mind that the most recent of the surveys at issue here is even now out of date by almost half a year.
Put in somewhat more legalistic language—even a non-lawyer cannot escape it!—it is my conclusion that additional severance and disclosure is possible, pursuant to section 25, without giving rise to the harm described in section 14. For that reason, I must find the complaints dealt with in this report to be well-founded. It is my recommendation that additional portions of the requested records be disclosed. A marked package of records indicating my proposed approach to satisfactory severance has been provided to your access coordinator.
The Information Commissioner determined that of the total of 709 pages all but 74 should be released.
C. RELEVANT STATUTORY PROVISIONS
Section 2 of the Access to Information Act sets forth the purpose of the Act:
2. (1) The purpose of this Act is to extend the present laws of Canada to provide a right of access to information in records under the control of a government institution in accordance with the principles that government information should be available to the public, that necessary exceptions to the right of access should be limited and specific and that decisions on the disclosure of government information should be reviewed independently of government.
(2) This Act is intended to complement and not replace existing procedures for access to government information and is not intended to limit in any way access to the type of government information that is normally available to the general public.
Subsection 4(1) [as am. by S.C. 1992, c. 1, s. 144] sets forth a basic right of access to government information. It states:
4. (1) Subject to this Act, but notwithstanding any other Act of Parliament, every person who is
(a) a Canadian citizen, or
(b) a permanent resident within the meaning of the Immigration Act,
has a right to and shall, on request, be given access to any record under the control of a government institution.
Sections 13 to 24 set forth exemptions from the general principle of access set out in subsection 4(1). Section 14 is the provision relied upon by the Government in this case for refusing to disclose the records requested:
14. The head of a government institution may refuse to disclose any record requested under this Act that contains information the disclosure of which could reasonably be expected to be injurious to the conduct by the Government of Canada of federal-provincial affairs, including, without restricting the generality of the foregoing, any such information
(a) on federal-provincial consultations or deliberations; or
(b) on strategy or tactics adopted or to be adopted by the Government of Canada relating to the conduct of federal-provincial affairs.
Section 25 provides for severance of disclosable from non-disclosable portions of the record:
25. Notwithstanding any other provision of this Act, where a request is made to a government institution for access to a record that the head of the institution is authorized to refuse to disclose under this Act by reason of information or other material contained in the record, the head of the institution shall disclose any part of the record that does not contain, and can reasonably be severed from any part that contains, any such information or material.
Sections 30 to 37 set forth the procedures involving the Information Commissioner.
Under section 41 persons who have been refused access to information, if they have made a complaint to the Information Commissioner, may apply to this Court for a review of the matter:
41. Any person who has been refused access to a record requested under this Act or a part thereof may, if a complaint has been made to the Information Commissioner in respect of the refusal, apply to the Court for a review of the matter within forty-five days after the time the results of an investigation of the complaint by the Information Commissioner are reported to the complainant under subsection 37(2) or within such further time as the Court may, either before or after the expiration of those forty-five days, fix or allow.
Under section 42 the Information Commissioner may apply to this Court for a review of any refusal to disclose in respect of which he has conducted an investigation provided he has the consent of the person who requested access:
42. (1) The Information Commissioner may
(a) apply to the Court, within the time limits prescribed by section 41, for a review of any refusal to disclose a record requested under this Act or a part thereof in respect of which an investigation has been carried out by the Information Commissioner, if the Commissioner has the consent of the person who requested access to the record;
(b) appear before the Court on behalf of any person who has applied for a review under section 41; or
(c) with leave of the Court, appear as a party to any review applied for under section 41 or 44.
Section 47 requires the Court, in any proceeding arising from applications under sections 41 or 42 to take every reasonable precaution to avoid disclosure of the record that is the subject-matter of the application. This unusual procedure was the subject of comment by Jerome A.C.J. in Maislin Industries Limited v. Minister for Industry, Trade and Commerce, [1984] 1 F.C. 939 (T.D.), at page 942:
On the subject of closed hearings, proceedings in our courts must take place in full public view and in the presence of all parties. Exceptions to this principle occur from time to time, but must be kept to the minimum of absolute necessity. Even then, directions should be such as to safeguard the public interest in the administration of justice, and the rights of any parties not permitted to participate. In applications under these access to information statutes, the issue is confidentiality, and obviously to conduct them in public view pre-empts the final decision. For the present, therefore, there does not seem to be any alternative but to restrict attendance to counsel for the parties.
In any proceeding pursuant to sections 41 or 42, section 48 places the burden on the Government to establish authorization to refuse disclosure:
48. In any proceedings before the Court arising from an application under section 41 or 42, the burden of establishing that the head of a government institution is authorized to refuse to disclose a record requested under this Act or a part thereof shall be on the government institution concerned.
Section 50 sets forth the role and jurisdiction of the Court in the case of an application arising out of the refusal by the Government to make disclosure based on section 14 of the Act:
50. Where the head of a government institution refuses to disclose a record requested under this Act or a part thereof on the basis of section 14 or 15 or paragraph 16(1)(c) or (d) or 18(d), the Court shall, if it determines that the head of the institution did not have reasonable grounds on which to refuse to disclose the record or part thereof, order the head of the institution to disclose the record or part thereof, subject to such conditions as the Court deems appropriate, to the person who requested access to the record, or shall make such other order as the Court deems appropriate.
D. THE LEGAL TEST FOR CONFIDENTIALITY
This is the first case in which this Court has been called upon to consider the exemption provided under section 14 of the Act (“injurious to the conduct by the Government of Canada of federal-provincial affairs”).
The Act contains two types of exemptions, mandatory (sections 13, 19, 20, 24) and discretionary (sections 14, 15, 16, 17, 18, 21, 22, 23). In the case of mandatory exemptions, the only decision to be made is whether the record comes within the description that the Act requires be exempted from disclosure. In the case of discretionary exemptions such as that under section 14, two decisions are necessary: first, does the record come within the description that is contemplated by the statutory exemption invoked in a particular case; and second, if it does, should the record nevertheless be disclosed. It is with respect to the first type of decision that evidence was filed and counsel devoted their arguments.
Subsection 2(1) which sets forth the purpose of the Act provides a guide to the interpretation of the operative provisions of the Act. It says that the purpose of the Act is:
2. (1) ... to extend the present laws of Canada to provide a right of access to information in records under the control of a government institution....
It establishes guiding principles:
(a) government information should be available to the public;
(b) necessary exemptions .... should be limited and specific;
(c) decisions on disclosure ... should be reviewed independent of government.
In considering the purpose of the Access to Information Act [S.C. 1980-81-82-83, c. 111, Sch. I] in Maislin Industries Limited v. Minister for Industry, Trade and Commerce, supra, Jerome A.C.J. stated at pages 942-943:
There was no disagreement that the burden of proof rests upon the applicant Maislin. It should be emphasized however, that since the basic principle of these statutes is to codify the right of public access to Government information two things follow: first, that such public access ought not be frustrated by the courts except upon the clearest grounds so that doubt ought to be resolved in favour of disclosure; second, the burden of persuasion must rest upon the party resisting disclosure whether, as in this case, it is the private corporation or citizen, or in other circumstances, the Government.
In this case the exception is claimed pursuant to section 14. The words “could reasonably be expected to” in section 14 are also found in other sections of the Act. In considering the same wording in paragraph 20(1)(c), the Federal Court of Appeal has determined that the exception to access must be based on a “reasonable expectation of probable harm”: Canada Packers Inc. v. Canada (Minister of Agriculture), [1989] 1 F.C. 47, at pages 59-60. A careful reading of the decision of MacGuigan J.A. reveals that he was “tempted” to construe the words “could reasonably be expected to” by analogy to an approach to tort law but resisted that temptation since it might open the door to an exception for possible rather than probable harm:
... I believe the temptation to carry through the tort analogy should be resisted, particularly if Wagon Mound (No. 2), supra, is thought of as opening the door to liability for the mere possibility of foreseeable damage, as opposed to its probability. The words-in-total-context approach to statutory interpretation which this Court has followed in Lor-Wes Contracting Ltd. v. The Queen, [1986] 1 F.C. 346; (1985), 60 N.R. 321 and Cashin v. Canadian Broadcasting Corporation, [1988] 3 F.C. 494 requires that we view the statutory language in these paragraphs in their total context, which must here mean particularly in the light of the purpose of the Act as set out in section 2. Subsection 2(1) provides a clear statement that the Act should be interpreted in the light of the principle that government information should be available to the public and that exceptions to the public’s right of access should be “limited and specific”. With such a mandate, I believe one must interpret the exceptions to access in paragraphs (c) and (d) to require a reasonable expectation of probable harm.
It is clear that the statutory provisions as they have been construed in the jurisprudence place a heavy burden upon the party seeking to prevent disclosure. Faced as they were with such an onus, counsel for the Government in this case attempted to persuade me that I should not follow Canada Packers, supra, and that I should have regard to Australian jurisprudence on the Australian Freedom of Information Act 1982 [No. 3, 1982] which is more favourable to the Government’s position.
The decision in Canada Packers, supra, is binding on me. Nonetheless, one of the Government’s arguments requires some brief comment. It was argued that subsection 2(1) of the Access to Information Act is merely descriptive and should not be used as a guide to interpreting the Act. I cannot accept this proposition. Statutes are to be interpreted with a view to carrying out their objects and purposes. When Parliament has been explicit in setting forth the purpose of an enactment and principles to be applied in construing it, I am of the opinion that such purpose and principles must form the foundation on which to interpret the operative provisions of the Act. Indeed, in the excerpt from the Canada Packers case, supra, quoted above, MacGuigan J.A. came to that very conclusion in respect of this Act.
Counsel for the Government also advanced arguments attempting to distinguish Canada Packers (it dealt with paragraph 20(1)(c) and not section 14) or attempting to argue that it was incorrectly decided. In pursuit of this argument, counsel also had to argue that Saint John Shipbuilding Ltd. v. Canada (Minister of Supply and Services) (1990), 67 D.L.R. (4th) 315 (F.C.A.), which followed Canada Packers was also wrongly decided. The words “could reasonably be expected to” as interpreted in Canada Packers and subsequently followed by Saint John Shipbuilding are the same as in this case. I do not consider it fruitful to dissect counsel’s arguments in detail since I am bound by Canada Packers and Saint John Shipbuilding in any event. Suffice it to say I have considered the arguments and am not persuaded by them.
As mentioned, the Government also submitted that guidance in interpreting and applying the Canadian Act may be found in examining the Australian jurisprudence as it relates to the Australian Freedom of Information Act 1982.
Subsection 3(1) of the Australian Act is a purpose clause similar but not the same as subsection 2(1) of the Canadian Access to Information Act. Subsection 3(1) of the Australian Act states:
3. (1) The object of this Act is to extend as far as possible the right of the Australian community to access to information in the possession of the Government of the Commonwealth by—
(a) making available to the public information about the operations of departments and public authorities and, in particular, ensuring that rules and practices affecting members of the public in their dealings with departments and public authorities are readily available to persons affected by those rules and practices; and
(b) creating a general right of access to information in documentary form in the possession of Ministers, departments and public authorities, limited only by exceptions and exemptions necessary for the protection of essential public interest and the private and business affairs of persons in respect of whom information is collected and held by departments and public authorities.
(2) It is the intention of the Parliament that the provisions of this Act shall be interpreted so as to further the object set out in sub-section (1) and that any discretions conferred by this Act shall be exercised as far as possible so as to facilitate and promote, promptly and at the lowest reasonable cost, the disclosure of information.
The Government submitted that the Australian equivalent to section 14 of the Canadian Act is found at subsection 33a(1) [as enacted by Freedom of Information Amendment Act, 1983, No. 81, s. 17] of the Australian Act which states:
33A (1) Subject to sub-section (5), a document is an exempt document if disclosure of the document under this Act—
(a) would, or could reasonably be expected to, cause damage to relations between the Commonwealth and a State; or
...
(5) This section does not apply to a document in respect of matter in the document the disclosure of which under this Act would, on balance, be in the public interest.
In support of its position, the Government referred to a number of Australian authorities. However, in argument, counsel submitted that Arnold (on behalf of Australians for Animals) v Queensland (1987), 73 ALR 607 (Aust. Fed. Ct.), provided a summary of the relevant Australian authorities and would be sufficient in outlining the interpretation he wished to advance.
In defining the proper test to apply under subsection 33A(1), Burchett J. in Arnold v. Queensland, supra, at page 628, wrote:
A feature of the drafting of para (a) of s 33A(1) is that it does not require a finding that disclosure would cause damage to relations between the Commonwealth and a State. It provides the alternative: “or could reasonably be expected to cause [such] damage”. It is now established that this does not require a probability, though a possibility which fails to reach the level of probability must be sufficiently tangible to answer to the notion of a reasonable capacity to be expected.
Later, on the same page the learned Judge stated:
I think the statute employs the expression “could reasonably be expected” in a sense corresponding to that which this passage conveys, but when all analysis has been exhausted the ultimate requirement of the statute is to apply its words directly to the problem thrown up by the circumstances of the case....
These two passages capture, I believe, the thrust of what the Government wished to import from the Australian jurisprudence. That is: (1) there should exist a sliding scale test where the more serious the injury from disclosure, the more one should accept a possibility of injury; the less serious the injury from disclosure, the more one would tip the balance toward requiring a probability of injury; and (2) the words of the Access to Information Act should be read on their own without the incorporation of other words into their meaning.
As enticing as these propositions may be, they simply cannot be accepted given the clear and direct Canadian authority in this area of the law. On the first point, the Canadian jurisprudence (Canada Packers, supra) precludes me from adopting a sliding scale evaluation that would justify confidentiality based upon a reasonable expectation of possible harm having regard to the seriousness of the injury that might be caused by disclosure.[4] On the second point, it is abundantly clear from the same jurisprudence that the test that I must apply in interpreting the words “could reasonably be expected to be injurious to” in section 14 of the Access to Information Act is one of reasonable expectation of probable harm.
The Canadian jurisprudence interpreting the Access to Information Act has established guidelines that can be useful in assessing whether or not there is a reasonable expectation of probable harm from disclosure in a given situation and the procedures to be followed. The following are not exhaustive:
1. The exceptions to access require a reasonable expectation of probable harm: Canada Packers, supra, at page 60.
2. The considered opinion of the Information Commissioner should not be ignored: Rubin v. Canada (Canada Mortgage and Housing Corp.), [1989] 1 F.C. 265 (C.A.), at page 272.
3. Use of the information is to be assumed in assessing whether its disclosure would give rise to a reasonable expectation of probable harm: Air Atonabee Ltd. v. Canada (Minister of Transport) (1989), 27 C.P.R. (3d) 180 (F.C.T.D.), at page 210.
4. It is relevant to consider if the information sought to be kept confidential is available from sources otherwise available by the public and whether it could be obtained by observation or independent study by a member of the public acting on his or her own: Air Atonabee, supra, at page 202.
5. Press coverage of a confidential record is relevant to the issue of expectation of probable harm from its disclosure: Canada Packers, supra, at page 63; Ottawa Football Club v. Canada (Minister of Fitness and Amateur Sports), [1989] 2 F.C. 480 (T.D.), at page 488.
6. Evidence of the period of time between the date of the confidential record and its disclosure is relevant: Ottawa Football Club, supra, at page 488.
7. Evidence that relates to consequences that could ensue from disclosure that describe the consequences in a general way falls short of meeting the burden of entitlement to an exemption from disclosure: Ottawa Football Club, supra, at page 488; Air Atonabee, supra, at page 211.
8. Each distinct record must be considered on its own and in the context of all the documents requested for release, as the total contents of the release are bound to have considerable bearing on the reasonable consequences of its disclosure: Canada Packers, supra, at page 64.
9. Section 25 of the Act provides for severance of material in a record that can be disclosed from that which is protected from disclosure under an exemption provision. The severance must be reasonable. To disclose a few lines out of context would be worthless: Montana Band of Indians v. Can. (Min. of Indian & Nor. Affairs), [1988] 5 W.W.R. 151 (F.C.T.D.), at page 166.
10. Exemptions from disclosure should be justified by affidavit evidence explaining clearly the rationale exempting each record: Ternette v. Canada (Solicitor General), [1992] 2 F.C. 75 (T.D.), at pages 109-110; and Merck Frosst Canada Inc. v. Canada (Department of Health and Welfare Protection Branch) (1988), 20 C.P.R. (3d) 177 (F.C.T.D.), at page 179.
In analyzing the material in this case I have had regard to these principles and procedures.
E. THE EVIDENCE
The evidence in these four motions consists of some sixteen affidavits and transcripts of cross-examinations on four of the affidavits. Each of the affidavits consists of voluminous exhibits and attachments. It would not be an exaggeration to say that the material in this case amounted to well over one thousand pages.
Evidence in Support of the Motions
1. Public Affidavit of Paul Tetro—June 16, 1992
Mr. Tetro is General Counsel in the Office of the Information Commissioner. In this affidavit Mr. Tetro recites the chronology of events relating to the Simpson and Lapointe access to information request. Attached as an exhibit are 108 pages (out of 709 pages) which the PCO indicated were the pages available for disclosure and which were severed from the balance of the information which it was alleged qualified for exemption under section 14 of the Act.
The affidavit explains that an investigator was appointed by the Information Commissioner’s Office, that all relevant pages, both confidential and non-confidential, were provided by the PCO to the investigator and that internal PCO memoranda relating to the access requests were also provided to the investigator.
The affidavit goes on to outline the procedure followed by the Information Commissioner to which I referred earlier.
The Information Commissioner process took from mid-January to mid-May, 1992. The affidavit concludes by explaining how the court applications were instituted.
2. Public Affidavit of Paul Tetro—July 22, 1992
This affidavit essentially covers the chronology of events relating to the access request of Nielsen and Morris. To all intents and purposes the facts outlined in Mr. Tetro’s affidavit of June 16, 1992 in relation to the access requests of Simpson and Lapointe apply to the access requests of Nielsen and Morris.
3. Confidential Affidavit of Paul Tetro—June 16, 1992
This affidavit lists all of the records identified by the PCO as being embraced by the original requests made pursuant to the Act and attaches all such records. It also refers to the access to information record of decision of the PCO, and internal PCO memoranda related to the access requests and attaches these documents. Various submissions and correspondence between the complainants and the Information Commissioner and the PCO and the Commissioner are attached. The documents recommended for disclosure are listed and attached and those not recommended for disclosure are attached.
4. Affidavit of Mary Calamai—June 23, 1992
This affidavit sets forth a chronology of events relating to Ms. Calamai’s request for access to information. Essentially the same course of events occurred as with the Simpson and Lapointe requests. Ms. Calamai deposes that the Information Commissioner advised that he would represent her interest in this Court but in doing so attached certain conditions. Ms. Calamai says that it is her firm belief that all of the records requested should be disclosed (and this is the reason she makes her own application and does not consent to be represented by the Information Commissioner). Attached to this affidavit are copies of the information requests, subsequent correspondence with the PCO and the Information Commissioner, the disclosed documents, some further information that was released at a later date, and notes from a 1984 address by the Prime Minister of Canada.
5. Affidavit of Mary Calamai—July 8, 1992
In this affidavit Ms. Calamai deposes that she is a researcher employed by Southam News which is a wire service owned and operated by Southam Inc. Southam News, together with the Angus Reid Group, conducts polls and surveys. Paragraph 4 of this affidavit states:
4. One area in which extensive polling has been conducted by the Angus Reid Group and by Southam News during the time period covered by my access request of November 7, 1991 (Exhibit “A” to my Affidavit of June 23, 1992) is with respect to national unity, the constitution and constitutional reform. Attached to this my Affidavit as Exhibits “A” to “M”, as more particularly listed below in paragraph 5, are polls that have been conducted by the Angus Reid Group and Southam News on the issues of national unity, the constitution and constitutional reform; the same issue that is the subject of these proceedings.
Paragraph 5 lists the names of 13 polls conducted from November 16, 1990 to May 7, 1992. Paragraphs 6 and 7 of this affidavit state:
6. Attached as Exhibits “1” through “60” of this my Affidavit, and contained in a separate bound volume, are computer printouts of articles from the same time period that have been written by reporters with Southam News with respect to these issues and these polls. As I am unaware of the content of those records that have not been released to me and which are the subject of these proceedings, I have produced all of the articles provided by Southam News during the relevant time period that touch on these issues. These articles have been made available for publication to newspapers across Canada by Southam News.
7. From review in the limited materials that were released to me by the Respondent to this Application, it is evident from pages 554 and 555 (as found in Exhibit “D” to my Affidavit of June 23, 1992), that two focus groups were held in which there was a “discussion on the Group of 22 proposals”, that participants were asked for “their reactions to this ‘federal proposal’” and that participants were asked to “read these 28 proposals.” Attached to this my Affidavit as Exhibit “61” is a copy of the Report of the Group of 22, including its 28 proposals and as Exhibit “62” are copies of newspaper articles reporting on reactions to these proposals.
Attached to this affidavit are the results of the 13 polls referred to in paragraph 5 and the 60 articles referred to in paragraph 7. In addition, attached is the “Report of the Group of 22”.
6. Affidavit of Ken Rubin—June 15, 1992
This affidavit sets forth a chronology of events commencing with Rubin’s access to information request dated November 29, 1991, the fact that the Information Commissioner offered to apply to the Federal Court on Rubin’s behalf but that Rubin “could not give him consent as requested to argue the case on my behalf because I wanted to receive the total 690 [sic] pages of the poll, not 620 [sic] pages.”
7. Supplementary Affidavit of Ken Rubin—June 29, 1992
This affidavit sets out a chronology of events and attached correspondence related to earlier access to information requests by Mr. Rubin. Also attached to this affidavit are contracts between the PCO and Decima Research Limited and Créatec+. These contracts are dated September 12, 1991 and set forth that the contracts are for public opinion research surveys as detailed in Appendix A for the period May 15, 1991 to September 30, 1991. The scope of work on Appendix A of each contract was not disclosed.
8. Affidavit of Ken Rubin—June 29, 1992
In this affidavit, Mr. Rubin deposes to the similarity of his access request with those of the Information Commissioner and Ms. Calamai. The affidavit was made in support of an application to join the Rubin proceedings with the others.
Evidence Opposing the Motions
9. Public Affidavit of Daniel Joseph Gagnier—July 6, 1992
During the period October 17, 1990 to June 30, 1992 Mr. Gagnier was Deputy Secretary to the Cabinet, Communications and Consultation, PCO. In paragraph 7 of his affidavit Mr. Gagnier deposes that he made the recommendation to the Deputy Head of the PCO, Mr. Tellier, to refuse to release records on the basis that disclosure could reasonably be expected to be injurious to the conduct by the Government of Canada of the issues concerning the Constitution and national unity. Paragraph 8 refers to the importance of the subject-matter:
8. These records relate to a subject that is unique and is of signal importance. That subject is the constitutional fabric of Canada. The conduct of the matter by the Government of Canada by this fact alone takes on a special significance—reflected as well in the nature and extent of the records.
Paragraphs 9 to 15 deal with the subject-matter of the record and why public opinion research is important to the Government. Paragraph 17 makes it clear that the refusal to release materials is related to the timing of the release:
17. It should be made clear that the refusal to release the materials is related to the timing of the release. It is not a refusal to release at some future time when the current stage of constitutional discussions has been completed and there is no reasonable expectation of injury to the conduct by the Government of matters relating to the constitution and national unity. The timing of release is related to the character of the records here in issue. Routine release of other types of records relating to public opinion studies within 90 days after a final report is received thereon is now under consideration. The remainder of this my affidavit addresses the expectation of injury from disclosure at this time of the particularly sensitive records in issue here.
Paragraph 20 outlines four general reasons for the Government’s refusal to release:
20. Four general reasons have been identified against a background of the acknowledged volatile public environment and the recognized object of the public opinion research—to assist the Government of Canada in its conduct of the important and intertwined constitutional and national unity issues. The four reasons can be described as:
1) constraining the strategic and tactical options available to the Government of Canada,
2) the potential negative impact on the environment for the federal-provincial negotiations,
3) prejudice to ongoing discussions with the provinces, and
4) furnishing information to those hostile to Canadian unity.
Paragraphs 21 to 23 set forth the background for the assessment of possible injury by the release of the records:
21. The assessment of possible injury by the release of the records must be dealt with against the background of the Government’s obligations in relation to constitutional and unity matters as part of its conduct of federal-provincial affairs having regard to:
a) the ongoing and increasingly urgent constitutional negotiations,
b) the difficult economic and social conditions reflecting on the perceived ability of the government to deal with unity matters,
c) the increasing complexity of these negotiations resulting from sovereignist sentiment in Québec, regional sentiment outside Québec and the demands of indigenous peoples and other groups, and
d) the possibility of referenda and elections which may alter the balance in the negotiations.
22. These factors result in a volatile public environment. There is also the background of the position in Confederation of the Province of Québec, the various problems engendered by, on the one hand, the refusal of the Québec Government to support the 1982 amendments to the Constitution and, on the other hand, the failure of the Meech Lake Accord. A companion concern is the attitude of residents of the other provinces and aboriginal groups to initiatives designed to deal with the perceived problem of constitutional legitimacy.
23. Finally, it should be emphasized that the polling data for which release has been sought was collected and analyzed for the sole object of assisting in the development and execution of the policies of the Government of Canada in relation to the conduct of matters respecting the amendment of the constitution and to the related subject of national unity.
Paragraphs 24 to 32 deal with the reasons why strategy and tactics of the Government could be compromised by disclosure:
Strategy and Tactics
24. Questions put and answers given in a public opinion survey or poll are likely to give an indication of what the Government might intend in terms of strategy and tactics in constitutional negotiations. Inferences will be drawn, whether correct or incorrect, and those inferences will have an effect on the positions and attitudes of other participants in the negotiations. The result will be that the negotiations will be faulted and prejudice to the Government’s capacity to carry out its responsibilities will occur. Specifically it will constrain its strategic and tactical options.
25. Among its various uses, public opinion research is undertaken to test the hypothetical proposals which will not necessarily be adopted. Release of such information could:
a) suggest to those with whom the Government of Canada has been dealing what it had been contemplating or, conversely, leave a false impression of its purposes, and
b) invite public pressure on both the Government of Canada and the provincial governments to either embrace or repudiate hypothetical proposals emerging from the research.
26. The information revealed to provincial negotiators, or indeed others at the negotiating table, by the release of polling results would, thus, do injury to the position in the negotiations of the Government of Canada by revealing its internal strategic and tactical deliberations. Revealing the information could do injury to the Government’s strategy and tactics by, for example, returning the agenda of intergovernmental deliberations to older issues. It could also overwhelm such deliberations with the interjection of new, manifold public pressures on a variety of hypothetical possibilities not now or ever intended to be central to current deliberations.
27. Responses to questionnaires could,
a) be used in the negotiations themselves by other parties, and
b) be reported by the media
in a manner that would be detrimental to the Government of Canada’s position.
28. To the extent that the media were to highlight findings that demonstrated lack of public support for a federal position, the federal negotiating position on that, and hence on other matters, could be undermined. To the extent that the media were to highlight findings that demonstrated profound regional divisions, the federal government’s ability to achieve inter-regional and inter-provincial conciliation could be injured. To the extent that the release of evolving polling data revealed when and why public preferences changed, other negotiators would acquire as much knowledge as the federal negotiators with respect to the sources and causes of opinion change on this vital matter. The Government of Canada would not have a similar advantage through access to the research of the provinces.
29. In this same regard, public opinion research done on a regular basis using the same questions, tracks the development of public opinion and selected segments of it. Revealing the questions could tend to leave a cloud over subsequent studies in that the results could not be relied on since they could reflect prior knowledge. More importantly, release would alert others involved in the negotiations to the Government’s ongoing concerns and, in the result, such polls might have to be abandoned. In consequence, the Government would be deprived of the benefit of knowing what public attitudes are on issues which it may select for tracking. Other parties to the negotiations would not be similarly handicapped.
30. Furthermore, a questionnaire, even one containing what may appear to be the most innocuous questions, is part of a complex matrix of polling questions the answers to which are used the one against the other to produce and verify information. Whether the questionnaire contains tracking questions or not, there is a danger of contamination for subsequent polls. At this critical juncture, the Government should not be deprived of the most accurate results possible, a consequence that may occur if questions are revealed.
31. The impact of public opinion on evolving negotiations is complex and dynamic. Regional differences in public opinion, for example, help to account for different provincial and federal negotiating positions. These regional differences are to some extent moderated over the course of extended negotiations. However, the release to the media of data evidencing substantial regional differences in public opinion would underscore those differences, and, hence, could dramatically reduce the ability of the parties to negotiations to reach compromise. Thus, the release of the Government of Canada’s data could seriously exacerbate inter-regional and inter-governmental tensions and constrain the Government of Canada’s strategic and tactical options. Furthermore, emerging from Government of Canada-sponsored research, such evidence would have more authority than, for example, evidence emerging from the work of an independent research or from privately sponsored polls.
32. The release of older public opinion research might appear at first blush to be less problematic than the release of current data. Older studies will have explored issues that have since been resolved or superseded in inter-governmental discussions. Unfortunately, the resurrection of these older issues in the media and in the public mind as if they were pressingly current may well divert the negotiations. The fact that media reporting of these results might specify the dates of these older findings would not diminish the injury. In the public mind, the older dates of these findings would be dwarfed in importance by the currency of media attention. In the mind of the general public, the importance of a phenomenon is often determined by whether or not the phenomenon receives current media attention, not by when the phenomenon actually occurred.
Paragraphs 33 to 35 deal with the negative impact on the environment for federal-provincial negotiations that would occur as a result of release of information:
Impact on Environment for Negotiations
33. The expected negative impact of the release of public opinion research on the environment in which constitutional review is taking place can be seen from what may be perceived to be divisions in Canadian society. Taking data secured at one point in time one might discern, for example, a dichotomy between public opinion in Québec and public opinion in the rest of Canada. Potential media reporting and third party exploitation of this kind of division could be very destructive at this critical juncture.
34. The general environment with respect to the constitution and national unity may also be affected adversely in that the communications approach of the Government, founded as it must be on securing the necessary compromise on the many divisive issues involved, would be prejudiced. If information in the records relating to public attitudes is revealed, it will be much harder to develop a persuasive and coherent communications approach that will bring Canadians together.
35. A related point is that the Government’s policy and its conduct of the matter may well be overtaken by peripheral or even false issues reported in the media which divert time and attention on the part of the Government from the matter at hand—the resolution of the current constitutional impasse. In other words, injury may be seen to the conduct of the file by reason of the release of information which may generate new issues not central to the debate which release, in turn, creates problems of response to the detriment of the conduct of the main issues. Moreover, public opinion studies lose their ability to assist in gauging public preferences and attitudes if the results become a major subject of public discussion themselves.
Paragraphs 36 to 39 deal with prejudice to ongoing discussions with the provinces:
36. The consequences of making other parties to the negotiations privy to information sought by the Government of Canada to assist, inter alia, in the conduct of those very negotiations cannot be underestimated. To know what the other party to the negotiation knows severely handicaps the party whose information is released. This is even more serious where the flow of information is in one direction.
37. In my experience as an adviser to the provincial government in constitutional negotiations (for Ontario inter alia in the Meech Lake negotiations), I would have felt it a distinct advantage for the provincial negotiators to have known what were the results of federal public opinion research and, particularly, what questions had been put. The advantage would have been increased by cross-checking information in provincial surveys on the same issue.
38. There is further potential injury in relation to negotiations in circumstances where a participant sees its constituency being surveyed by another participant. This is especially true concerning issues which may be of particular importance to such a participant. In my experience this might well create an atmosphere little conducive to moderation and compromise thereby making the Government of Canada’s task that much more difficult.
39. Potential negotiating problems as a result of release abound. If public opinion studies were released, it could create controversy and undermine the flexibility the Government of Canada has sought to maintain having regard both to the importance of the matter and the diversity of views which pertain.
Paragraphs 40 to 42 deal with furnishing information to those hostile to Canadian unity:
Information to Those Hostile to Canadian Unity
40. It is the Government of Canada’s responsibility to promote a resolution and to bring all of the participants in the constitutional and national unity debate together. Release of the public opinion research may be expected to be detrimental to the Government’s catalytic role and, hence, to the conduct of the matter.
41. It may be anticipated that public opinion research, if released, whether it is reported or not, will find its way into the hands of those who will seek to use it for purposes totally at variance with what must be the object of the Government of Canada’s conduct of the matter.
42. This information, if released, could also be used to promote the ends of those hostile to Canadian unity in an indirect fashion. Currently the attention and resources of the Government of Canada in this area are fully committed. Resources are finite and it will be a serious detraction from the imperative task of the moment for the Government of Canada to have to deal with the contentious use that one may expect to be made in the public arena of the research materials by those seeking to promote disunity.
10. Confidential Affidavit of Daniel Joseph Gagnier—July 6, 1992
In this affidavit Mr. Gagnier sets out illustrations from the confidential records of the various propositions put forward in his public affidavit.
11. Public Affidavit of Frederick James Fletcher—July 6, 1992
Dr. Fletcher is professor of political science at York University. In addition to his other experience, Dr. Fletcher is Research Coordinator, Media and Elections, for the Royal Commission on Electoral Reform and Party Financing. Dr. Fletcher was asked to advise the Government of Canada with respect to the effect of the release of the public opinion research which is the subject of these motions. Dr. Fletcher attaches a report to his affidavit (Exhibit B) covering his examination of the matter.
Dr. Fletcher first provides comments on background and context. For example:
2. In considering the release of the public opinion studies in question here, it is necessary to place them in the context of federal-provincial affairs in general and of the Government of Canada’s conduct of the national unity file. The Government of Canada has a unique responsibility for national unity and for protecting the legitimacy of the constitution.
3. Constitutional negotiations are particularly difficult because the issues are fundamental and abstract. The negotiations have important symbolic content that is exacerbated when regional and linguistic tensions are involved. The mobilization of groups of citizens around symbolic issues makes already difficult negotiations more difficult (Simeon and Robinson, 1990:323) [sic].
...
9. The decision of the Privy Council Office to propose that public opinion studies on the national unity issue be made an exception to the developing practice of releasing government public opinion studies relatively soon after their completion must be understood, in my view, in this context. The unity negotiations are fundamental, involve substantial risks to the future of the country and are currently at a critical juncture. The environment for these negotiations is volatile and the issues are complex. In short, the sensitivity of these negotiations is at a very high level.
...
12. In the context of national unity, the Government of Canada has a responsibility to strive to keep the country together and, therefore, a duty to seek consensus and to avoid polarization. To discharge these duties effectively, the Government needs the information provided by public opinion studies. Such studies allow policy-makers and negotiators to take public opinion into account in developing positions, formulating negotiating strategies, and selecting the time and form of public communication and consultation. The struggle for control of the public agenda is a central process in federal-provincial negotiations, especially when symbolic issues are involved. “In this struggle,” as a leading public opinion researcher has observed, “the many kinds of support or opposition revealed in public opinion surveys are valuable resources” (Johnston, 1986:217).
Paragraphs 21 to 25 of his report fall under the heading “Media Use of Public Opinion Studies”:
21. Public opinion studies are technical documents. In order to interpret them properly, it is necessary to understand such matters as probability sampling, basic social statistics, the influence of question wording and order effects, at the least, as well as the context in which individual studies were conducted. The reports provided by survey firms frequently assume at least some aspects of this knowledge by clients and do not always provide it.
22. Although some political journalists have the necessary qualifications and experience, many do not. This often leads to a failure to understand the technical limitations of polls and to misinterpretation of results. The Royal Commission on Electoral Reform and Party Financing (Canada 1991:458-9) documented the potential for such misinterpretation. Scholarly studies have noted others, including one that could have affected the outcome of a federal party leadership convention (Fletcher, 1988:102). The distinguished journalist Jeffrey Simpson (1987:6) summed up the risks of misinterpretation when journalists report on public opinion studies:
Many journalists, including those who write about political matters, are unschooled in Canadian history, in polling methodology, in an understanding of any part of the country but their own. Yet the beguiling simplicity and easy accessibility of polling data embolden all journalists to become instant pundits, or worse still, experts. They can pontificate on the meaning of this, the likely outcome of that, the significance of everything, on the basis of a few stark numbers.
Pollsters regularly comment on the lack of technical competence of journalists in dealing with public opinion studies (see, for example, Round Table, 1991:16).
23. The focus of the news media is on generating news; i.e., stories of interest to their audiences. Most news organizations, therefore, define news as that which is dramatic, surprising or important. David Taras (1988:222), one of Canada’s leading experts on news coverage of political issues, comments on news practices in the major Canadian media as follows:
Drama is the prime ingredient of most news stories. Drama requires sensationalism, winners and losers, and ... most of all ... conflict. As a CBC news editor observed ...’ “We look for conflict often to the exclusion of the story” ’.
In the context of Canadian federal-provincial relations, information that emphasizes conflict, especially regional and linguistic conflict, is regarded by the news media as highly newsworthy. It is viewed as dramatic and important. Media reporting of lengthy and complex materials, such as the public opinion studies in question here, is inevitably partial and selective and likely to reflect standard news practices. After careful study of television news coverage of the debate surrounding the Meech Lake Accord, Taras (1988:222) concluded that news reporting, especially on television, presents conflict in a “point-counterpoint format.” Although many issues cannot easily “be collapsed into a choice between pro and con”, news reports frequently force information into this two-sided format to heighten the sense of conflict. It seems clear that the probability of the public opinion studies in question here being reported in this way adds to the likelihood that the release of the records would make compromise more difficult and therefore [sic] to the difficulty faced by the Federal Government in conducting the national unity file.
24. In reporting public opinion studies, information that highlights regional or linguistic conflict (or ethnic conflict) is likely to be stressed, along with other findings that are surprising or controversial. Few news organizations are in a position to report the materials in their full context and complexity.
25. In the context of these records, two of the consequences of the inevitable selective reporting might be (1) to return issues to the constitutional agenda that had been thought to be resolved or put aside and/or (2) to focus attention on substantively minor but sensationalized items. Both of these eventualities would distract attention from more important issues and divert the resources of the Government of Canada from central to more peripheral issues. The resources of the Federal Government, especially the time and energy of senior officials, are not unlimited.
Commencing at paragraph 26, Dr. Fletcher addresses the question of “Dangers of Releasing the Public Opinion Studies”. He first argues that the studies should be seen as integrated studies that, taken as a whole, carry a risk of damage if released (paragraph 29). He says that it is a common but unfortunate error to view public opinion studies as individual commodities rather than as parts of an ongoing process of research, analysis and more research (paragraph 30).
In general, Dr. Fletcher says that research sponsored by the Government of Canada will have greater credibility and therefore a greater potential to influence public perceptions than other public opinion studies (paragraph 31). He expresses concern that in future, knowledge that public opinion studies might be released while a file is still active might well inhibit the use of some kinds of questions—such as scenarios or other hypothetical questions subject to misinterpretation—thus reducing the utility of public opinion studies for policy-makers (paragraph 32).
Paragraphs 33 to 37 describe constraints on the strategic and tactical options of the Government of Canada. Paragraphs 38 to 44 deal with negative impacts on the environment of federal-provincial constitutional negotiations. Paragraph 45 deals with the possibility of damage to ongoing negotiations with the provinces:
45....
As noted above, the selective reporting of dramatic or unusual features of the findings might well return an issue to the already overloaded constitutional agenda. This would set back the negotiations and constitute a drain on the limited resources of the negotiators. The latter consequence would be particularly the case if the data were reported in a misleading manner or misinterpreted by the media or by other interested parties. Considerable time could be lost in attempting to correct misperceptions at a critical time in the negotiations.
In paragraphs 46 to 50, Dr. Fletcher addresses the question, “At what point are public opinion studies suitable for release?” He says the mere passage of time does not necessarily remove the potential for damage—that some issues are likely to remain sensitive until the negotiations are complete. In paragraph 47 he states:
47....
The fact that sensitive material is from older polls is not likely to mitigate potential damage. The influence on the general public of particular findings will be determined primarily by the amount of attention accorded them in the major news media. Disclaimers about the timing of a particular study—or the political context within which it was conducted—are not likely to make much of an impression. The capacity of media attention to place a topic on the public agenda is well-documented as is the tendency of the public to recall broad impressions rather than specific details from news coverage.
In paragraph 51, Dr. Fletcher addresses the subject “Furnishing Information to Those Hostile to National Unity”:
51. As noted above, constitutional negotiations in practice involve many more participants than those with official roles. Unofficial participants have considerable capacity to mobilize public sentiment and to bring pressure to bear on negotiators. The government of Canada public opinion studies would have utility to all of those groups. They would be especially useful to groups that wish to denigrate Federal Government efforts or to identify issues where potential exists to mobilize sentiments hostile to the success of the negotiations. There are, in my view, some data in the records that could be used to oppose compromise.
At paragraph 52 Dr. Fletcher concludes:
52. It is my view that the release of the public opinion studies would constrain the strategic and tactical options available to the Government of Canada, could raise false issues that would distract attention from more substantive concerns, possibly weaken the bargaining position of the Federal Government, might encourage division in the country, and would likely reduce the effectiveness of future public opinion studies for the Government of Canada.
12. Confidential Affidavit of Frederick James Fletcher—July 6, 1992
In this affidavit Dr. Fletcher refers to examples in the confidential information that illustrate some of the propositions contained in the report attached to his public affidavit.
Affidavits in Rebuttal
13. Supplementary Affidavit of Allison Lawford—July 8, 1992
Allison Lawford is a research assistant in the office of the Information Commissioner. Attached as exhibits to her affidavit are approximately 100 press clippings related to public opinion polls, poll reports, focus groups and focus group reports on constitutional matters, published from April 1990 to June 29, 1992.
14. Supplementary Confidential Affidavit of Paul Tetro—July 8, 1992
This affidavit refers to access to information requests (other than those giving rise to these motions) to the PCO and correspondence and documents relating thereto. Mr. Tetro deposes that these documents are relevant because the documents released pursuant to these other applications are polls or reports on polls dealing with constitutional or national unity issues. In particular, Mr. Tetro states that one document which the PCO refused to disclose pursuant to the requests of Simpson and Lapointe was disclosed pursuant to a previous request.
15. Public Affidavit of Richard Johnston—July 8, 1992
Dr. Johnston is professor of political science at the University of British Columbia. Among other things, Dr. Johnston is a co-author of Letting the People Decide: Dynamics of a Canadian Election. That book deals with, among other things, the impact of polls and other media factors on election campaigns. Dr. Johnston was asked to provide his opinion with respect to the opinions given and conclusions drawn by Mr. Gagnier and Dr. Fletcher. In paragraph 6, Dr. Johnston provides an overview of his opinion:
6. Based on my professional experience and knowledge of the theory and practice of public opinion research, it is my opinion that:
(a) the public release of the records attached to the Confidential Affidavit of Paul B. Tetro, taking into account their age and content, is most unlikely to have measurable effects on public attitudes;
(b) in particular, the academic literature of which I am aware does not support the proposition that the public knowledge of the results of surveys of opinion on public issues could be reasonably expected to result in polarization of opinion, a “bandwagon effect”, or a so called “spiral of silence” effect; and
(c) the academic literature of which I am aware also does not support the proposition that disclosure of results of surveys of opinion on public issues could reasonably be expected to contaminate future “tracking studies” on similar issues.
In the balance of his affidavit, he provides detailed argument in support of his opinion.
16. Supplementary Confidential Affidavit of Richard Johnston—July 8, 1992
In this affidavit Dr. Johnston identifies tracking data in the record and comments on it.
Cross-examination on Affidavits
Mr. Gagnier and Dr. Fletcher were cross-examined on their public and confidential affidavits by counsel for the Information Commissioner, counsel for Ms. Calamai and by Mr. Rubin.
17. Cross-examination of Daniel Joseph Gagnier—Public Affidavit—July 8, 1992
The examination included the following subject-matters:
(a) information that has already been released—questions 111-113;
(b) concern over the media’s use of polls—question 144;
(c) whether or not there is a document-by-document analysis of the record—questions 163-167;
(d) the PCO both refusing and agreeing to release the same poll (C-4 in the record)—question 197;
(e) whether harm resulted from the release of the C-4 poll—questions 198-205;
(f) harm to the government strategy and tactics—questions 204-221;
(g) concerns over the way in which the media could use the information as opposed to “ordinary Canadians”—questions 250-256.
18. Cross-examination of Daniel Joseph Gagnier Confidential Affidavit—July 8, 1992
This cross-examination dealt with specific polling results. The associated questions and answers covered whether the polling information was already known to Canadians, whether the subject-matter would come as an unusual surprise to Canadians, the way in which stale information might be dealt with, and the similarity of some polls to those published by other organizations.
19. Cross-examination of Frederick James Fletcher—Public Affidavit—July 8, 1992
This cross-examination dealt with the following matters among others:
(a) the probability versus possibility of harm from disclosure—questions 43-44;
(b) the merits of disclosure—questions 79-85 and 97;
(c) whether or not a page-by-page analysis of the record was conducted by Dr. Fletcher—question 101;
(d) the nature of the media to report in a divisive manner—questions 190-191;
(e) the basis of considering whether the media would draw attention to division and conflict—question 205;
(f) tracking and the future utility of polls—questions 216-221;
(g) identifying Government strategy—questions 222-228;
(h) the position of the Royal Commission on Electoral Reform and Campaign Financing on the question of publicizing Government polls—questions 231-233.
20. Cross-examination of Frederick James Fletcher Confidential Affidavit (Now Public)—July 8, 1992
This cross-examination included matters such as:
(a) the degree of detail in which Dr. Fletcher considered the polling information—questions 247-249;
(b) the use of polling information by the media—question 255;
(c) the detail with which Dr. Fletcher reviewed each of the 74 pages which the Information Commissioner did not ask be released—questions 265, 268, 269.
F. THE CHARTER OF RIGHTS
Counsel for the Information Commissioner and Mr. Rubin relied on the Access to Information Act in support of their applications. Counsel for Ms. Calamai in support of her application primarily relied upon the Canadian Charter of Rights and Freedoms, being Part I of the Constitution Act, 1982, Schedule B, Canada Act 1982, 1982, c. 11 (U.K.) [R.S.C., 1985, Appendix II, No. 44], and in particular paragraph 2(b) which guarantees:
2. ...
(b) freedom of thought, belief, opinion and expression, including freedom of the press and other media of communication;
In response to the invocation of the Charter, counsel for the Government brought to the Court’s attention section 57 of the Federal Court Act [R.S.C., 1985, c. F-7 (as am. by S.C. 1990, c. 8, s. 19)]. Subsection 57(1) states:
57. (1) Where the constitutional validity, applicability or operability of an Act of Parliament or of the legislature of any province, or of regulations thereunder, is in question before the Court or a federal board, commission or other tribunal, other than a service tribunal within the meaning of the National Defence Act, the Act or regulation shall not be adjudged to be invalid, inapplicable or inoperable unless notice has been served on the Attorney General of Canada and the attorney general of each province in accordance with subsection (2).
Subsection 57(2) [as am. idem] states:
57....
(2) Except where otherwise ordered by the Court or the federal board, commission or other tribunal, the notice referred to in subsection (1) shall be served at least ten days before the day on which the constitutional question described in that subsection is to be argued.
The hearing of the argument commenced on August 11, 1992. Ms. Calamai’s factum was served on the Government on August 4, 1992 (as was agreed to by the parties). Thus the ten-day notice period required for service on the Attorney General of Canada was not met. No notice at all was served on the attorney general of each province.
Counsel for Ms. Calamai took the position that section 57 was not applicable in this case because Ms. Calamai was not questioning the constitutional validity, applicability or operability of section 14 (or any other section) of the Access to Information Act. Rather, and it was put variously during argument, counsel argued:
1. section 14 must be interpreted or construed in light of paragraph 2(b) of the Charter;
2. the interpretation of section 14 is to be narrowed by use of the Charter;
3. paragraph 2(b) of the Charter and section 14 of the Act can be argued in parallel. The existence of section 14 does not affect the qualified right of access to information under paragraph 2(b). Where disclosure of information would contribute to core values, a prima facie right of access is protected by paragraph 2(b);
4. for a government official to focus on what the media will do with information was to take account of an extraneous, irrelevant or improper factor constituting an infringement of paragraph 2(b).
Counsel pointed out that no Canadian court has yet extended the scope of the freedoms protected under paragraph 2(b) to access to government information. Counsel established that freedom of expression in the context of the press not only includes the right to disseminate information but also the right to gather it. Authority was cited in which the freedom to gather information was said to recognize the right of the press to be present at various judicial and even quasi-judicial proceedings. Counsel urged that it was not a giant leap of faith to find that paragraph 2(b) applied to the gathering of government information of the type under consideration in this case.
I find the arguments involving the Charter to be complex indeed. For the purposes of my analysis I will assume, without deciding, that paragraph 2(b) of the Charter applies to the type of government information under consideration in this case. I now address the question of whether this is an attack on the constitutional validity, applicability or operability of section 14, requiring compliance with section 57 of the Federal Court Act.
We are here dealing with a decision of the Prime Minister of Canada as head of the PCO, to disclose certain documents and withhold others pursuant to section 14 of the Access to Information Act. (In practice an elaborate decision-making process is involved in which a recommended decision is sent up through the ranks of the PCO, is verified as being consistent with the Access to Information Act by legal counsel, and is verified as to all appropriate administrative actions having been taken. At this point a “final decision” is made by an Assistant Secretary of the PCO and the Deputy Head of the PCO.)
In considering how the Charter might apply in this case I am guided by the analysis of Lamer J. (now C.J.C.) in Slaight Communications Inc. v. Davidson, [1989] 1 S.C.R. 1038. Although in that case what was at issue was an order of an adjudicator made under the Canada Labour Code [R.S.C. 1970, c. L-1], the analysis of Lamer J. is in my view applicable to the case at bar in assessing the nature of the decision of the head of the government institution to refuse disclosure made under the Access to Information Act. I quote from the reasons of Lamer J. at pages 1077 to 1080:
The fact that the Charter applies to the order made by the adjudicator in the case at bar is not, in my opinion, open to question. The adjudicator is a statutory creature: he is appointed pursuant to a legislative provision and derives all his powers from the statute. As the Constitution is the supreme law of Canada and any law that is inconsistent with its provisions is, to the extent of the inconsistency, of no force or effect, it is impossible to interpret legislation conferring discretion as conferring a power to infringe the Charter, unless, of course, that power is expressly conferred or necessarily implied. Such an interpretation would require us to declare the legislation to be of no force or effect, unless it could be justified under s. 1. Although this Court must not add anything to legislation or delete anything from it in order to make it consistent with the Charter, there is no doubt in my mind that it should also not interpret legislation that is open to more than one interpretation so as to make it inconsistent with the Charter and hence of no force or effect. Legislation conferring an imprecise discretion must therefore be interpreted as not allowing the Charter rights to be infringed. Accordingly, an adjudicator exercising delegated powers does not have the power to make an order that would result in an infringement of the Charter, and he exceeds his jurisdiction if he does so. This idea was very well expressed by Professor Hogg when he wrote in his text titled Constitutional Law of Canada (2nd ed. 1985), at p. 671:
The reference in s. 32 to the “Parliament” and a “legislature” make clear that the Charter operates as a limitation on the powers of those legislative bodies. Any statute enacted by either Parliament or a Legislature which is inconsistent with the Charter will be outside the power of (ultra vires) the enacting body and will be invalid. It follows that any body exercising statutory authority, for example, the Governor in Council or Lieutenant Governor in Council, ministers, officials, municipalities, school boards, universities, administrative tribunals and police officers, is also bound by the Charter. Action taken under statutory authority is valid only if it is within the scope of that authority. Since neither Parliament nor a Legislature can itself pass a law in breach of the Charter, neither body can authorize action which would be in breach of the Charter. Thus, the limitations on statutory authority which are imposed by the Charter will flow down the chain of statutory authority and apply to regulations, by-laws, orders, decisions and all other action (whether legislative, administrative or judicial) which depends for its validity on statutory authority.
Section 61.5(9)(c) must therefore be interpreted as conferring on the adjudicator a power to require the employer to do any other thing that it is equitable to require the employer to do in order to remedy or counteract any consequence of the dismissal, provided however that such an order, if it limits a protected right or freedom, only does so within reasonable limits that can be demonstrably justified in a free and democratic society. It is only if the limitation on a right or freedom is not kept within reasonable and justifiable limits that one can speak of an infringement of the Charter. The Charter does not provide an absolute guarantee of the rights and freedoms mentioned in it. What it guarantees is the right to have such rights and freedoms subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society. There is thus no reason not to ascribe to Parliament an intent to limit a right or freedom mentioned in the Charter or to allow a protected right or freedom to be limited when the language used by Parliament suggests this.
It would be useful, in my view, to describe the steps that must be taken to determine the validity of an order made by an administrative tribunal, which are as follows:
First, there are two important principles that must be borne in mind:
— an administrative tribunal may not exceed the jurisdiction it has by statute; and
— it must be presumed that legislation conferring an imprecise discretion does not confer the power to infringe the Charter unless that power is conferred expressly or by necessary implication.
The application of these two principles to the exercise of a discretion leads to one of the following two situations:
1. The disputed order was made pursuant to legislation which confers, either expressly or by necessary implication, the power to infringe a protected right.
— It is then necessary to subject the legislation to the test set out in s. 1 by ascertaining whether it constitutes a reasonable limit that can be demonstrably justified in a free and democratic society.
2. The legislation pursuant to which the administrative tribunal made the disputed order confers an imprecise discretion and does not confer, either expressly or by necessary implication, the power to limit the rights guaranteed by the Charter.
— It is then necessary to subject the order made to the test set out in s. 1 by ascertaining whether it constitutes a reasonable limit that can be demonstrably justified in a free and democratic society;
— if it is not thus justified, the administrative tribunal has necessarily exceeded its jurisdiction;
— if it is thus justified, on the other hand, then the administrative tribunal has acted within its jurisdiction.
In the context of the analysis of Lamer J. in Slaight Communications and considering the Charter challenge raised in this case, section 14 of the Access to Information Act:
(1) confers either expressly or by necessary implication the power on the head of the government institution to infringe an assumed protected right; or
(2) confers imprecise discretion and does not confer, either expressly or by necessary implication, the power on the head of the government institution to limit the rights guaranteed by the Charter.
If the decision by the head of a government institution to refuse disclosure is made pursuant to express or implied power granted under section 14, a Charter challenge to section 14 would be involved.
If the decision by the head of a government institution to refuse disclosure is made pursuant to a broad discretionary power, then the decision itself is subject to a Charter challenge. Such a Charter challenge would not require compliance with section 57 of the Federal Court Act because the validity, operability or applicability of a statute is not being questioned.
In the case of the Access to Information Act, the exceptions to disclosure are intended to be limited and specific. The decision under section 14 is one which requires assessment of whether disclosure could reasonably be expected to be injurious to the conduct of federal-provincial affairs by the Government of Canada. This is not a broad discretionary decision of the type under consideration in Slaight Communications, supra, at page 1079, where the adjudicator could order an employer to:
... do any other thing that it is equitable to require the employer to do in order to remedy or counteract any consequence of the dismissal ....
In my view the decision to be made under section 14 is confined to the formulation of an opinion as to whether or not disclosure of information could reasonably be expected to be injurious. While section 14 does not exhaustively list the considerations to be taken into account in the decision-making process, there is no doubt that it does nothing other than empower the making of a decision that documents fall into a category exempt from the general rule of disclosure and permits confidentiality if they do. In my opinion, this is a situation analogous to the first situation identified by Lamer J. in Slaight Communications, supra. If access to government information is a protected right then it is section 14 that expressly confers the power to limit that right. As such, a Charter challenge such as the one in this case must be directed against the legislation.
Counsel for Ms. Calamai says that he does not challenge the constitutionality of section 14. Indeed he says it is valid and there are occasions when a refusal to disclose would be justified. However, if section 14 expressly confers the power on the government to limit an assumed protected right (access to government information), the attack must be on the constitutional validity, applicability or operability of section 14. This requires compliance with section 57 of the Federal Court Act.
In an effort to get around section 57, counsel argues that section 14 must only be “construed” with a view to paragraph 2(b) of the Charter and that this is different than questioning the validity, applicability or operability of the section.
While there may be circumstances where an argument relating to the construction of a statute does not involve the question of its validity, applicability or operability, I cannot appreciate such a distinction in this case based upon the arguments made. Counsel argues that the information in question here contributes to “core values” thereby creating a prima facie right of access and that in these circumstances the exemption in section 14 is narrowed by paragraph 2(b) of the Charter. This argument if accepted would, to my mind, result in the inapplicability or inoperability of the exemption under section 14 or at least the limiting or narrowing of the applicability or operability of the exemption when documents relating to core values are at issue. If it does not result in the limiting or narrowing of the applicability or operability of the exemption then “construing” section 14 in light of the Charter serves no useful purpose.
With respect to the Charter arguments, therefore I conclude:
1. The Charter challenge necessarily involves a challenge to section 14 of the Access to Information Act.
2. To the extent section 14 was found to contravene paragraph 2(b) of the Charter the result would be its inapplicability or inoperability or at least a limitation on its applicability or operability. Such a determination requires notice in accordance with section 57 of the Federal Court Act which was not given.
Accordingly, in the absence of required notice, which may possibly lead to the hearing of additional submission, I will not adjudicate the Charter challenge in this case.
Having said this, it may not be inappropriate for me to comment on the argument that in this case the decision to refuse disclosure was based upon what the media “will do” with the information and that this was an extraneous, irrelevant or improper factor to take into account, thereby putting the decision outside paragraph 2(b) of the Charter. The duty of the government official is to form an opinion as to whether disclosure could reasonably be expected to be injurious to the Government’s position in the conduct of federal-provincial affairs. It is one thing to argue that there is no expectation of probable injury from use of information by the media. It is another to argue that a decision is patently unreasonable or that extraneous, irrelevant or improper factors have been taken into account because regard is had to what the press might do with the information. I am not aware of any case which has held that consideration of the press’ handling of information is something extraneous to an assessment of probable injury. Indeed, in both Canada Packers, supra and Ottawa Football Club, supra, what the press might do with information was a consideration.
G. ANALYSIS BY THE COURT
1. The Released Pages
On January 8, 1992, 108 pages were released to all the requesters. An additional 12 pages were released to Mr. Rubin in respect of his specific access request. Thus as of January 8, 1992, 120 of 709 relevant pages had been released.
Of these 120 pages, 57 could be described as introductory or cover pages. For example, they state:
[p. 447]:
TOP LINE QUALITATIVE REPORT
SIGMUND QUEBEC GENERAL PUBLIC
15/10/91
| Prepared for: |
FPRO Garry Breen |
| Prepared by: |
Simon Corneille John Paterson |
| Date: |
October 22, 1991 |
| [p. 448]: |
INTRODUCTION
• This report presents the results of two focus groups held in Montreal on October 15, 1991. Given that the week was relatively calm in political terms, we can report no significant recent event that may have influenced the reactions of participants.
• The reader is reminded of the qualitative, as opposed to quantitative nature of this study. As such the results reported herein should not and cannot be construed as representative of the larger population in any statistically significant manner.
A further 42 pages consist of pages of what appear to be general categorization questions asking respondents in what age category they belong, their primary language, citizenship, province of residence, sex, etc.
The 12 pages provided in response to Mr. Rubin’s request covered the contracts entered into for polling by the Government, with the exception of the scope of work portions on 2 pages which were not disclosed.
A third category of released pages consists of pages which describe the methodology or approach in the conduct of the polls or summarize the demographic breakdowns of the persons participating in the focus groups. These numbered fourteen. An example of the former would be page 484:
OBJECTIVES
[s. 14]
METHODOLOGY
• Two sessions were held in Montreal in which a total of 18 French speaking Quebequers participated. Each session lasted approximately one hour and forty-five minutes. Participants in both groups were recruited according to the following standard criteria:
— 1/2 men, 1/2 women, all Francophones, among a broad range of ages;
— people with diverse occupations
— [section 14]
An example of the latter would be page 614.
PROJECT: 412-002
| [TRANSLATION] |
July 16, 1991 6 p.m. Montréal |
July 16, 1991 8 p.m. Montréal |
| • Men • Women |
6 6 |
6 6 |
| • 18 - 23 • 24 - 39 • 40 - 54 |
2 5 5 |
2 5 5 |
| • Students • Housewives • Blue-collar workers • White-collar workers • Unemployed or jobless |
1 (2) 4 4 1 |
1 (2) 4 4 1 |
| • Mother tongue: French |
12 |
12 |
| [section 14] • Undecided |
0 - 3 |
0 - 3 |
As indicated by the reference to “section 14” on both pages 484 and 614, there were portions that were not disclosed.
On June 18, 1992 a further 35 pages were released by the Government. One was an introductory page. Eight pages consisted of results of a poll or survey done on or about November 15, 1990 on the Citizens’ Forum (Spicer Commission). The questions asked about a respondent’s familiarity with the Citizens’ Forum, whether the respondent approved of the Forum, whether the Forum was independent from government, why a respondent thought the Forum was set up, what the expected performance of the Forum was, and whether a respondent agreed with different comments on the Government’s decision to set up the Forum. From the correspondence between the Information Commissioner and the PCO it appears that at first the Government was of the view that these pages might not be relevant to the access request in this case. After re-examination they were released but no explanation is given as to why the Government decided to do so. An additional seven pages consisted of general questions relating to classifications by age, income, etc. Finally, a further 19 pages were released because the Government came to the conclusion that they were duplicates of material which was previously released (presumably in response to other access requests).
In total, prior to the start of this hearing, 155 pages had been released. Of this total, 29 pages were released in part only, e.g. 484, 614 referred to above.
The subject-matter of most of the pages that were released could fairly be described as being generic or neutral in the sense that they do not contain information about constitutional or national unity issues. Where pages do make reference to constitutional or national unity issues they appear to have been released because they were released previously pursuant to another access request. All other pages were kept confidential by the Government.
2. The Confidential Pages
Of the 709 pages of the record that were relevant, the Government seeks to maintain confidentiality over 554 pages. Without revealing the content of these confidential pages, I can give some indication of their nature.
The first and largest category consists of 222 pages containing questions which provide the “raw data” of the polls. A question is posed followed by a percentage breakdown of the responses. The structure of these pages are typical of a poll one would find in a newspaper or magazine. These pages do not contain any conclusory or analytical statements.
A second category of 55 pages consists of analysis of the “raw data”. Here conclusions are drawn based upon the data collected. As well, some analysis of trends is provided.
There are 132 pages providing cross-tabulations and demographic analysis of polling data. These are highly complex and very technical documents which are of little practical use to the uneducated observer.
There are 45 pages that describe the questions which were asked of the various focus groups that were held. These appear to be the structure or framework that the leaders of the focus groups used to direct the participants of the groups to the specific topics to be canvassed.
The analysis of the focus groups covers some 83 pages. Here conclusions are provided regarding the reaction of the participants of the groups to selective topics that were raised. In addition to the analysis of these reactions, some recommendations based on the results of the groups is also provided.
There are two pages that generally describe the objectives of the focus groups.
There are 14 pages that formally set out the position on constitutional matters of a group of individuals who met for the express purpose of formulating proposals regarding the constitution.
Finally, there is one page that defies classification. It appears to be an option chart of some kind.
3. The Role of the Court under Section 50
In considering the analysis to be done by this Court in a case under section 50 of the Access to Information Act, some guidance may be obtained by comparing the role of the Court with that of the Information Commissioner. As I perceive it, although there are some similarities, the role of the Information Commissioner is somewhat different than that of the Court. The Office of the Information Commissioner is an office which apparently deals with access to information requests on a regular basis. There may be extensive dialogue that takes place, apparently informally, between the Information Commissioner and his staff, and the government department that is the subject of an access request. For example, in this case, in his January 16, 1992 letter to the Clerk of the Privy Council, the Information Commissioner states:
... and taking into account the extensive discussions our offices have already had about the accessibility of unity polls, I am expediting the investigation ....
As well, Mr. Gagnier in his affidavit refers to numerous meetings between the Information Commissioner’s Office and the PCO.
The investigation by the Information Commissioner is conducted privately. The Information Commissioner does not appear to be bound by the ordinary rules of evidence. He may hear the parties separately. It is obvious in this case that the Information Commissioner analyzed the record on a page-by-page and indeed line-by-line basis.
By contrast, the Court’s process under section 50 relies on evidence. Evidence is necessary to link a reasonable expectation of probable harm to disclosure of specific information. The Court is not an expert tribunal with its own resources. It can only act on the basis of the evidence before it. In the absence of evidence, the Court has no basis or expertise of its own for developing its view as to whether or not the government department had reasonable grounds to refuse disclosure. In short, the heavy onus placed on the party seeking to maintain confidentiality must be satisfied in a formal manner on a balance of probabilities through clear and direct evidence. I will have more to say about this later.
4. The Presumption in Favour of Disclosure v. an Overly Cautious Approach to Disclosure
The Access to Information Act and the jurisprudence cited earlier makes it abundantly clear that Parliament’s intention is that access to government information is a right of all citizens and permanent residents and that exceptions are to be limited and specific. The Access to Information Act did not simply confirm in statutory form the law that existed before it came into force in respect of information under government control. On the contrary, in respect of certain government information such as that relating to federal-provincial relations, the law prior to the Access to Information Act gave the Government the absolute right to maintain confidentiality even against the normal Court discovery process in litigation.
Subsection 41(2) of the Federal Court Act, R.S.C. 1970 (2nd Supp.), c. 10, stated:
41....
(2) When a Minister of the Crown certifies to any Court by affidavit that the production or discovery of a document or its contents would be injurious to international relations, national defence or security, or to federal-provincial relations, or that it would disclose a confidence of the Queen’s Privy Council for Canada, discovery and production shall be refused without any examination of the document by the court.
Upon the Court receiving an affidavit of a Minister certifying that disclosure would be injurious to federal-provincial relations, no explanation was required in order to justify confidentiality. The Court was obliged to refuse disclosure without any examination of the documents in question. There was no testing of the Minister’s opinion. There was no consideration of whether disclosure would indeed be injurious or whether a Minister was simply adopting an unduly cautious approach.
With the enactment of the Access to Information Act in 1982, section 41 of the Federal Court Act was repealed (S.C. 1980-81-82-83, c. 111, s. 3). Parliament had demonstrably swung the pendulum away from absolute confidentiality which the Government enjoyed in areas such as federal-provincial relations and toward disclosure of all documents under Government control with only narrow and limited exceptions.
The contrast between the repealed subsection 41(2) of the Federal Court Act and the provisions of the Access to Information Act reveals that Parliament was no longer satisfied with unjustifiably cautious approaches by Government to disclosure. If confidentiality of information was to be maintained, the onus was on the Government (or other party seeking to maintain confidentiality of information under Government control) to justify it by demonstrating that disclosure could reasonably be expected to be injurious.
The focus in this case, based on the evidence, is whether the Government had reasonable grounds to expect harm from disclosure or whether it was taking an overly cautious approach to disclosure, based on possibility, which is not justified by the legislation.
5. The Evidentiary Requirement
The jurisprudence indicates that the Government or party seeking to maintain confidentiality must demonstrate its case clearly and directly. The Act itself, in subsection 2(1), states that exemptions from disclosure must be limited and specific. By inference I think it is clear that a general approach to justifying confidentiality is not envisaged.
I assume, from the evidence of Mr. Gagnier, that in the absence of direction to the contrary, many conscientious public servants, in good faith, would opt for confidentiality when they perceived the slightest possibility of risk to Government initiatives from disclosure of information. He put it clearly in answer to question 137 of his cross-examination:
I think both the Information Commissioner, as well as myself, and maybe others, went through an assessment of that risk. Where the difference lies is he made his assessment judging some of the material to carry an unacceptable [sic] risk; whereas, I, probably being a bureaucrat and more cautious, judge that risk in this unique period of time was unacceptable. [Emphasis mine.]
Indeed, such an approach was permitted by a provision such as the now repealed section 41 of the Federal Court Act. However, today, an overly cautious approach, based on something less than a reasonable expectation of probable harm, is not to be consistent with the test under section 14 of the Access to Information Act.
In order to distinguish between confidentiality justified by the Act and that resulting from an overly cautious approach, specific detailed evidence is required.
In the case at bar, what is being considered is the validity of an opinion of a government official that disclosure of specific government documents could reasonably be expected to result in harm to the Government’s conduct of federal-provincial affairs. While no general rules as to the sufficiency of evidence in a section 14 case can be laid down, what the Court is looking for is support for the honestly held but perhaps subjective opinions of the Government witnesses based on general references to the record. Descriptions of possible harm, even in substantial detail, are insufficient in themselves. At the least, there must be a clear and direct linkage between the disclosure of specific information and the harm alleged. The Court must be given an explanation of how or why the harm alleged would result from disclosure of specific information. If it is self-evident as to how and why harm would result from disclosure, little explanation need be given. Where inferences must be drawn, or it is not clear, more explanation would be required. The more specific and substantiated the evidence, the stronger the case for confidentiality. The more general the evidence, the more difficult it would be for a court to be satisfied as to the linkage between disclosure of particular documents and the harm alleged.
In addition, allegations of harm from disclosure must be considered in light of all relevant circumstances. In particular, this includes the extent to which the same or similar information that is sought to be kept confidential is already in the public realm. While the fact that the same or similar information is public is not necessarily conclusive of the question of whether or not there is a reasonable expectation of harm from disclosure of the information sought to be kept confidential, the burden of justifying confidentiality would, in such circumstances, be more difficult to satisfy.
6. The Approach of Mr. Gagnier and Dr. Fletcher
Mr. Gagnier in his public affidavit refers to the unique importance and special significance of the subject-matter of the record in this case. He then sets out four reasons justifying confidentiality of the record:
1. Constraining the strategic and tactical options available to the Government of Canada.
2. The potential negative impact on the environment for federal-provincial negotiations.
3. Prejudice to ongoing discussions with the provinces.
4. Furnishing information to those hostile to Canadian unity.
In the balance of his affidavit he explains in some detail his concerns. He says that disclosure of the Government’s strategy or tactics in negotiations, even if such strategy could only be inferred from the information would constrain the Government’s options and could compromise its strategy. He expresses concern about the highlighting of conflict by the media and the creation of divisiveness in the country that would be counter-productive to the Government’s interest in seeking consensus in its constitutional and national unity initiatives. He says that misunderstandings may arise from disclosed information which may be exploited to the detriment of the Government’s position. He also refers to the contamination of subsequent polling results by premature disclosure of prior results thus rendering polling of reduced value to the Government.
Dr. Fletcher speaks to the same issues—the compromising of Government strategy and tactics, misinterpretation of the results of disclosed information, the highlighting of conflict by the media leading to increased polarization of public opinion and the contamination of subsequent tracking polls.
These are all reasons why the Government might, but not necessarily probably would, suffer harm from the disclosure of its confidential information. I can even appreciate the argument made with respect to misunderstandings in this case even though the jurisprudence tends to discount this type of concern because it is normally within the power of the party forced to release the record to provide additional information to clarify misunderstandings. If it could be demonstrated that considering all the circumstances, it is reasonable to expect the harm described from the release of the specific information in question, then the Government would be entitled to confidentiality. This requires evidence linking the harm described and the disclosure of specific pages of the record and an explanation of why, in all the circumstances, the disclosure of the contents of the record would cause such harm.
In paragraph 21 of his public affidavit, Mr. Gagnier says that one of the background considerations is:
... the increasing complexity of these negotiations resulting from sovereignist sentiment in Quebec, regional sentiment outside Quebec and the demands of indigenous people and other groups.
This is significant. It is apparent that Mr. Gagnier’s position is not that topics such as these are confidential. These subject-matters have been disclosed in the public material filed by the Government. If harm is to be found it must be in the disclosure of the details in the record and the affidavit evidence must link harm to disclosure of those details.
Language in the affidavits of Mr. Gagnier and Dr. Fletcher is often couched in the terminology of section 14. However, the evidence must demonstrate a probability of harm from disclosure and not just a well-intentioned but unjustifiably cautious approach to the avoidance of any risk whatsoever because of the sensitivity of the matters at issue.
Although both Mr. Gagnier and Dr. Fletcher state they are of the opinion that disclosure of these documents would probably result in harm to the position of the Government, the question remains: Are those views in harmony with the preponderance of probabilities which can readily be found in the evidence?
With respect to Mr. Gagnier’s evidence, he deposes that he was the one who made the original recommendation to maintain confidentiality over the documents in question. It will be remembered that the only pages which he agreed should be disclosed originally were introductory and cover pages, pages containing generic questions and pages disclosing certain methodology and demographic summaries. No other pages were disclosed. It appears to me that Mr. Gagnier, mindful of the “unique and … signal importance” and “special significance” of the subject-matter and “being a bureaucrat” took a very cautious approach.
While a cautious approach may not, in every case, be inconsistent with what is required under section 14, I observe that in this case Mr. Gagnier chose to disclose nothing of a substantive nature except where a prior disclosure had already been made. In preparing his affidavit he did not do a page-by-page analysis of the record to determine if specific pages could be released. In cross-examination, there appears the following:
166 Q. So what you have done now to advance it; you still haven’t done an analysis document-by-document. You have given some examples in your confidential affidavit?
A. That’s correct.
167 Q. So there is no recorded document-by-document analysis with respect to these documents and the section 14 exemption that exists to this point in the record.
A. Not in the record.
His approach suggests that his intention was to avoid any risk to the Government initiative, probable or possible. Indeed in paragraph 21 of his affidavit he states: “The assessment of possible injury by the release of the records” [emphasis mine].
In addition, in his cross-examination, Mr. Gagnier expresses a concern about the media misconstruing the Government polls. After making reference to a particular issue, the cross-examination continues:
501. Q. Now, I will stop you right there. That means you are afraid the media will not accurately report these polls.
A. Could.
Q. Might not; could not?
A. Might not.
Q. They might not and that is the danger you are talking about here.
A. I am not saying they would do it maliciously.
Q. Of course not. I have been careful to make sure you are not saying that but you are saying they might.
A. Yes.
While I accept that Mr. Gagnier expresses a serious concern about this issue, his evidence is that this “might” occur. However section 14, as it has been interpreted, requires evidence of probable harm. That the media “might” not accurately report polls does not equate to a probability that this will occur.
I conclude that while his motive was undoubtedly well intentioned and honest, and based on a serious concern of what might or possibly would occur, Mr. Gagnier was not, on a specific page-by-page basis, guided by the requirements of section 14.
With respect to Dr. Fletcher, he expressed his view as to the appropriate approach to disclosure in a case under the Access to Information Act in his cross-examination.
43 Q ... Would you agree with me that the phrase “may reasonably be expected to occur” could refer to the possibility of something happening or the probability of something happening, or, perhaps, somewhere between those two points on the continuum of likelihood. Would you agree with me?
A. I would, yes.
44 Q. At the time you swore this affidavit what was your understanding of where that phrase “may reasonably be expected to occur” fell on that continuum.
A. My earlier answer, I think, wasn’t unresponsive to your question. It seems to me that the more critical the situation, the more sensitive the information, the more one would accept some degree of possibility of injury. The less sensitive, the less critical the context, the more one would look to the probable level.
45 Q. So your understanding was that it was a point on the continuum which varied in accordance with the possibility of harm at the other end. Is that a fair way of putting it?
A. That’s right. [Emphasis mine.]
Dr. Fletcher’s approach seems to be closer to the Australian approach which accepts the possibility of injury as long as it is sufficiently tangible to be expected rather than the Canadian approach which requires a probability of injury. His is a “sliding scale” approach whereby if the consequences of disclosure are very serious, possible rather than probable harm should be accepted to justify confidentiality. In fact, at paragraph 29 of his public affidavit, he goes so far as to say:
The fact that a proposal has been publicly presented or discussed does not necessarily eliminate the possibility of damage ... [Emphasis mine.]
In this context, at paragraph 9 of his public affidavit, Dr. Fletcher makes clear how critical he views the consequences in this case:
The unity negotiations are fundamental, involve substantial risks to the future of the country and are currently at a critical juncture. The environment for these negotiations is volatile and the issues are complex. In short, the sensitivity of these negotiations is at a very high level.
His own words suggest that Dr. Fletcher is concerned with eliminating the possibility of damage. This, however, is not the test under section 14.
Dr. Fletcher admitted in cross-examination that he did not do a question-by-question analysis of the record:
247 Q. Can you say that you have read each and every question and answer that is set out in the materials.
A. No.
Apparently from the analysis he did do, he did not find it appropriate to recommend the disclosure of even one page of the record. This is surprising since, in support of his assertions of harm, he cites two pages (pages 28 and 36) that, apparently unbeknown to him, had previously been released by the Government on June 18, 1992. (Dr. Fletcher simply cites these pages in support of concerns about harm that might occur.) Further, the references in his confidential affidavit to those portions of the record that support the assertions in his public affidavit are often references to large blocks of pages. There is little in the way of specific detailed reference to individual pages (less than a dozen pages including pages 28 and 36 are referred to individually) or indeed individual questions.
Again in the case of Dr. Fletcher, I conclude he used a very cautious approach to disclosure guided by his “sliding scale” approach to the issue. In view of his concerns over the sensitive nature of the subject-matter in this case, his rather general approach to the record, the fact that he did not find even one page that could be disclosed, and his own words in his affidavit, it appears that Dr. Fletcher was prepared to accept possibility rather than probability of harm in order to justify confidentiality.
I indicated earlier that if the information which the Government sought to keep confidential is already in the public realm, the Government would find it more difficult to justify confidentiality. The jurisprudence indicates, and it stands to reason, that once information is public from another source, the release of the same or similar information by the Government will be less likely to cause harm. If there were harm from disclosure, that harm could reasonably be expected to have arisen from the prior disclosure by others. In such circumstances the Government would have to show specific reasons why its release of the same information would cause harm.
I believe it is significant that with one minor exception, Mr. Gagnier, in his affidavits makes virtually no reference to prior disclosure by others (he makes reference in paragraph 31 of his public affidavit to the fact that Government sponsored polls would have more authority than those of others). That Mr. Gagnier did not deal with this aspect in his affidavit causes me to have doubt that in forming his view that harm could be expected from disclosure of Government information, he took into account all relevant circumstances. In my view, an expectation, not based on all available and relevant information, is not the reasonable expectation called for by section 14.
In the case of Dr. Fletcher, in his affidavit, he speaks to the issue of other polling information being public, but does not deal with this in terms of specific comparisons. Indeed in his cross-examination he was asked:
52 Q. At any time did you look at the survey results in opinion polls other than the opinion polls in issue here which, in fact, have been released.
A. I did not look at those, no.
...
83 Q. You would agree with me, would you, that if these polls were released that the public would then be able to know what other people think about these issues?
A. Yes. They would make a contribution to that knowledge, or they might. I would have to look very carefully at what other polling information has been available previously to make a final determination of that matter.
Notwithstanding those answers, he appears to have considered overlap of Government and public polling information in a cursory manner:
75 Q. How long was this morning’s conversation with her?
A. Mainly we were sitting in the same room doing other work but there was a brief discussion of an affidavit that had arrived last night or this morning—
76 Q. From Mr. Johnston?
A. No, from Mary Calamai.
77 Q. In the conversation with Ms. Binns this morning did you discuss any information which was related to any of the matters that you have opined on in your Affidavit?
A. I don’t think so. The matter that I did ask her about was the extent to which the material in the Angus-Reid polls and the Affidavit overlapped with the materials which we are discussing here.
78 Q. Did you consider that question of how much the overlap was?
A. Yes.
However, that consideration must have been general in nature because Dr. Fletcher admitted that he did not make specific comparisons of the public and Government polls. This leads me to question how an effective comparison of the overlap could have been made.
As I concluded in respect of Mr. Gagnier, Dr. Fletcher’s approach causes me to have concern that he also did not adequately take into account all relevant circumstances in forming his view that harm that could reasonably be expected from disclosure of confidential information in this case.
Dr. Fletcher discounts the importance of other polling information being public in an assessment of whether harm could be expected by the release of Government polls. He and Mr. Gagnier assert that Government polls have more authority or credibility than those of others. In particular, Dr. Fletcher says that “it seems likely” that they will have more potential for influencing public perceptions than other polls. No other opinions are cited in support of these assertions. Dr. Johnston, in his affidavit, took the opposite view. While I have no doubt that the views of Mr. Gagnier and Dr. Fletcher are honestly held, without some further basis of support, this evidence is, in my view, impressionistic or subjective and of insufficient weight to satisfy the burden on the Government on this point.
In paragraphs 48 to 50 of his public affidavit, Dr. Fletcher asserts that disclosure of Government-commissioned polls, even if other polls may have been published, involves other special risks. He says that such risks involve disclosure of the Government’s negotiating strategy. This would be the case if the confidential record contained negotiating strategy or if such strategy could be inferred from the polling or focus group information itself. Nothing in Dr. Fletcher’s material demonstrates that this is the case. Mr. Gagnier, in cross-examination, admitted that the confidential record did not contain strategic or tactical information and that the inferences that others might draw from the information about the Government strategy would be inconclusive (questions 209 to 221).
Without reference to the information itself and an explanation of how release of specific pages or questions would disclose, directly or indirectly, Government strategy, or otherwise harm the Government, I cannot conclude that the evidence on this point is anything more than speculative.
Dr. Fletcher also asserts that certain parties might not approve of some questions being asked by the Government. However, again no reference to specific questions was made by Dr. Fletcher in support of this argument. While caution in this area is justified, I have no way of assessing whether there is a reasonable expectation of harm without being provided with information as to the parties that might be offended, the questions that might offend them, why they would be offended and the effect of similar questions having been asked in other public polls.
In cross-examination, (e.g. questions 494 to 496), Mr. Gagnier expressed concern that release of government information could lead to divisiveness, even though the same subject-matter was canvassed in public polls. While I accept that the release of information that would probably lead to divisiveness would be harmful to the Government’s position, it was not explained why this would occur in the face of the same or similar information having been the subject of public polls. Mr. Gagnier was also of the opinion that release of information in the record that was an updated version of information that was the same as that which had been previously released by the Government at a prior point in time would be harmful. Again it was not explained why, in such circumstances, harm would result.
Dr. Fletcher says that the wording in the Government polls might be different than those in other polls. I accept that this is true. He did not indicate the differences in wording that formed the basis of his concern nor why the specific wording in Government polls as opposed to that in public polls would probably result in harm if disclosed. In his cross-examination (e.g. questions 384 to 393), Mr. Gagnier referred to different questions on the same subject-matter as between the Government and public polls. However, no explanation was given as to why, if indeed a Government poll question was different from a public poll question, disclosure of the Government poll would be injurious.
Perhaps because of their view that extreme caution was warranted, Mr. Gagnier and Dr. Fletcher did not consider it necessary to compare the Government information with information made public by others and explain their views specifically in this context. In my opinion, this was a relevant and essential consideration in the formulation of an opinion as to whether or not harm could reasonably be expected from disclosure of the Government information in this case.
7. Specific References to Pages in the Record
Mr. Gagnier’s and Dr. Fletcher’s affidavits refer specifically to 56[5] pages in the record and to 182[6] pages only generally. No mention is made of the balance of 323 pages.
I reviewed the 56 pages referred to specifically. These pages consist of polling and focus group results and the scope of work portions of the Decima and Créatec+ contracts. In the absence of public information, including polling information, similar to the Government information, there may well have been reasonable grounds to justify confidentiality of some of these pages. However, the public information must be taken into account. While some of the public information only came into existence after the initial refusal to disclose by the PCO in January 1992, I am of the opinion that all such information is relevant. It was open to the Government, based on new information about public polls and public information, to change its mind up to the time of the hearing.
I compared the specific polling and focus group pages which the Government asserts should be kept confidential to published polls and press articles submitted in the evidence of the applicants to determine the extent to which disclosure of the Government commissioned polls would result in new information being made public. It would not be appropriate for me to disclose the details of the analysis because by doing so I would inadvertently be disclosing the subject-matter and results of the Government-commissioned polls contrary to section 47 of the Act. However, in partial satisfaction of the desirability to explain the basis of my findings, I will briefly outline the approach used.
I first reviewed the specific provisions of Mr. Gagnier’s and Dr. Fletcher’s public affidavits and the corresponding paragraphs of their confidential affidavits. I next reviewed the pages in the record specifically referred to in the confidential affidavits to ascertain the linkage between the harm alleged in the public affidavits and disclosure of these specific pages in the record. It was at this point that in some cases a prima facie case for confidentiality became apparent.
I then reviewed the evidence submitted by the applicants consisting of public polls and press articles relating to polls to determine if there was similarity between them and the specific pages Mr. Gagnier and Dr. Fletcher sought to keep confidential. The following is a table setting forth the analysis made. I have not referred to the specific public polls or press articles by page number because to do so would reveal, by comparison, the material the Government seeks to keep confidential. As an alternative, I have indicated the number of public references I was able to locate that were the same as or similar to the specific references to Government information in Mr. Gagnier’s and Dr. Fletcher’s confidential affidavits.
[Endnote for the *7[7]]
There are a total of six questions (two of which are duplicates) in the pages referred to specifically by the Government to which I cannot find corresponding questions in the material submitted by the applicants. However, the subject-matter of two of these questions did not appear to be strongly connected to the theme of constitutional reform and national unity. The four subject-matters covered by these questions all appear to have been dealt with within public polls and press articles, at least peripherally. Taken on their own, and without further explanation as to how harm would probably occur by their release, I am not convinced that confidentiality is justified.
While the questions in some of the public and Government polls are identical, more often, a public poll was merely similar but not identical with those commissioned by the Government. There is a high degree of overlap. As I indicated earlier, nothing in the material presented by the Government indicates to me that where the subject-matter is same, the specific wording of a Government poll is so substantially different from a public poll that a negotiating strategy or other sensitive information would be newly disclosed or that other harm would likely result from release of the Government information.
The only other specific reference made by Mr. Gagnier to documents that should be kept confidential was with respect to the scope of work portions on two pages of consulting contracts let to Créatec+ and Decima Research Ltd. I have reviewed the details contained in the scope of work portions of these contracts.
In the cross-examination on his affidavit Mr. Gagnier states in resisting disclosure of these pages:
Q. 654 ... why can’t you get it?
A. Because the scope provides an insight into the parameters which the Government contracted the research firm to investigate and might be amended over time to either add to that field or to drop certain parts from it.
Q. 655 But would you not agree that the work is already completed so why would that still be a concern?
A. It is ongoing. Parts of that scope of work, in the original contract that was let, might be reflected in subsequent contracts. Many times the contracts are simply extended. You would have to ask Federal Provincial Relations Office which they have done. I don’t know.
I do not see in Mr. Gagnier’s answers any rationale that would enable me to conclude that release of these pages would cause harm to the Government in its federal-provincial negotiations. The fact that the scope of work portion of the contracts provides an insight into the parameters for which the Government contracted does not of itself suggest that harm would befall the Government from its disclosure. That the scope might be amended over time does not relate to the potential for harm to the Government.
I am unable to conclude that there were reasonable grounds to refuse disclosure of these 56 specifically referred to pages.
8. General References to Pages in the Record
I have already noted the following: in their affidavits and cross-examinations, Mr. Gagnier and Dr. Fletcher referred in total to 238 pages in the record. Fifty-six were referred to specifically and those have been dealt with above. The balance of 182 pages are referred to only generally or in large blocks of pages by the witnesses. Four of these pages had been disclosed voluntarily by the Government—two on January 8, 1992 and two on June 18, 1992.
I will deal now with those 182 pages that are dealt with only generally. (The following paragraphs refer to 226 pages. The number 182 is net of 44 pages that were also dealt with specifically.)
A desirable procedure that has been found helpful is to set out on each page for which exemption from disclosure is sought, the specific injurious effect the release of that page would be likely to cause. In Ternette v. Canada (Solicitor General), [1992] 2 F.C. 75 (T.D.), MacKay J. states at pages 109-110:
I note for the record that the supplementary affidavit filed in confidence on behalf of the respondent did fully set out, with examples, the concerns about specific injurious effects anticipated to the interests of CSIS and efforts toward detecting subversive or hostile activities, if the information were released. In the exhibits to that affidavit which included all information not released to the applicant, including all pages which had been released with deletions, on each page was noted, by reference, the specific injurious effect or effects that release of that page was anticipated to be likely to cause. That thorough, careful analysis and documentation made it possible for the Court to review without difficulty the basis upon which the decision had been made to refuse access or release to the applicant.
I am not in that position with respect to these 182 pages of the record in this case. I do not have before me a page-by-page description of the harm that would be probable from disclosure of each page. In the case of these pages in the record to which only general reference has been made in the affidavits, the deponents have, to all intents and purposes, left the documents to speak for themselves as to how they are linked to the arguments made in the public affidavits and why their disclosure could result in harm to the Government. I have reviewed the pages to which general reference is made. I note that the Information Commissioner, who apparently did have extensive meetings and discussions with the PCO with respect to all pages, determined that all these 182 pages should be disclosed.
The pages consist of reports on polling results, polling results themselves and reports on the results of focus groups on a variety of subjects in the national unity and constitutional reform discussions. The pages in this generally referred to category are as follows:
1. Paragraph 7 of the confidential affidavit of Mr. Gagnier refers to Tabs C-5 to C-11, at pages 47-138, some 92 pages, as examples in support of paragraph 23 of his public affidavit which states:
23. Finally, it should be emphasized that the polling data for which release has been sought was collected and analyzed for the sole object of assisting in the development and execution of the policies of the Government of Canada in relation to the conduct of matters respecting the amendment of the Constitution and to the related subject of national unity.
I have no doubt that as deposed to, these polls were commissioned to assist the Government in the development and execution of policies in respect of the Constitution and national unity, but without an explained linkage of a described harm to disclosure of the specific information contained on the relevant pages, I cannot conclude that there were reasonable grounds on which to refuse to disclose them.
2. Paragraph 12 of the confidential affidavit of Mr. Gagnier refers to Tab C-14, at pages 202-310, as an example in support of paragraph 30 of his public affidavit. This material refers to cross-tabulations of results. Paragraph 12 of the confidential affidavit states (and I set it out because, other than the coded reference, it says essentially the same thing as paragraph 30 of the public affidavit):
With respect to paragraph 30, examination of the extensive analysis contained in C-14, pp. 202-310 will show how results are cross-tabulated. The detailed breakdown shows why release of any of the research information could be injurious and why, if subsequent polls were contaminated, the utility of the research might well be significantly diminished.
In respect of the cross-tabulation of the results, Professor Fletcher in his cross-examination was asked (question 248):
Q. How much time have you spent reviewing the material and I am talking about reviewing the material which is contained in the confidential affidavit of Mr. Tetro?
A. I spent approximately three days looking at them but the cross-tabulations, for example, you can spend an hour trying to understand one, but I have looked at most of the questions and the distribution of opinion on most of them.
No detailed explanation of cross-tabulations were provided. As Professor Fletcher says, “you can spend an hour trying to understand one”.
This general reference covers some 109 pages. Indeed the cross-tabulations are not explained and, as indicated by Professor Fletcher, they are not easy to understand. The Government has not satisfied me, on the basis of the evidence, that there were reasonable grounds to refuse disclosure of these pages.
3. Paragraph 7 of Professor Fletcher’s confidential affidavit refers to portions of the record in support of paragraph 28 of his public affidavit. Paragraph 28 of his public affidavit states:
While the 70 [sic] pages identified by the Information Commissioner as inappropriate for release clearly carry a high probability of damage, other substantive materials in the records also involve similar risks.
Paragraph 7 of the confidential affidavit says in part (and I cite only those portions which disclose the references):
With respect to paragraph 28, there are the detailed discussions of (two specific issues in the constitutional reform and national unity discussions). Among many possible examples, C-5, pp. 50-59; C-7 ( ); C-10 ( ); C-12 and much of the focus group material (C-18 ff.).
C-5, C-7 and C-10 were referred to by Mr. Gagnier and have been dealt with above. C-12 and C-18 cover some 25 pages. It is apparent from paragraph 28 of his public affidavit and paragraph 7 of his confidential affidavit, that the references to the record by Dr. Fletcher are very general. There is an overall concern about the risk of disclosure but nothing of any specific nature to assist me to understand why release of specific pages would result in harm. Again I cannot conclude that there were reasonable grounds to refuse to disclose the pages referred to by Dr. Fletcher in paragraph 7 of his confidential affidavit.
Without belabouring the point, in their confidential affidavits, Mr. Gagnier and Dr. Fletcher made reference to subsets of these same blocks of pages in relation to other paragraphs of their public affidavits. There were 11 references. The subsets range from 5 to 39 pages. Essentially, as in the three examples detailed above, the pages are left to speak for themselves. As I have indicated, this is insufficient in my view to establish a right to confidentiality in a complex matter such as the one before me.
In the absence of detailed and understandable explanations as to how and why release of each of the 182 pages which were referred to only generally could reasonably be expected to cause harm, the Government, in my view, has failed to discharge the burden placed on it by section 48 of the Act.
9. Pages not Referred to in the Affidavits of Mr. Gagnier and Dr. Fletcher
Three hundred and twenty-three confidential pages have not been referred to at all in the affidavits of Mr. Gagnier and Dr. Fletcher. I expect that it was their intention that all pages should be considered under an “incorporation by reference” approach. Indeed Dr. Fletcher argues that the studies done by the Government in this case should be seen as integrated and considered as a whole. While I accept the documents must be considered in context and in terms of how they relate to one another, this does not justify a general all-encompassing approach to the exclusion of consideration of the record on a page-by-page basis.[8]. Nor should the Court be left to analyze complex documents on its own. The Court is not a specialized body with its own experts.
While no specific evidence was led by the Government on these pages, evidence was led by the applicants with respect to some of them.
The information found at pages 576 to 589 of the confidential record is exactly the same as that to which the Court was referred by counsel for Ms. Calamai in the public evidence of the applicants. Mr. Gagnier admitted on cross-examination that these pages were “already public information” (question 599).
A further 17 pages all contain the same descriptive phrase. These are pages 450, 460, 484, 511, 532, 555, 592, 611, 613, 614, 637, 652, 655, 665, 679, 682 and 684. That phrase placed focus group respondents in a specific category which led to their being chosen for the focus groups. This same phrase is contained in a group of documents that was released by the Government pursuant to an access request unrelated to those before me in this application. The 17 pages of the record are referable to various focus groups conducted between May and October 1991. The released material which contains this same phrase is dated within that time period and refers to focus groups. In argument, counsel for the Government stated that the released material containing this phrase was different from the non-released pages which contain the same phrase in that it did not reveal the methodology of the focus groups and was simply a factual statement. However, the subject (focus groups) and timing (mid-1991) of the public information leads me to believe that this phrase is already in the public domain in the same context in which it appears in the 17 pages.
In response to a request by the Court, counsel for the Information Commissioner prepared an example of a comparison of information which was in the public realm which he said was similar to the confidential Government information. I have reviewed that table of comparisons and I am satisfied with the similarity of the material.
The foregoing causes me to conclude that in respect of these 323 pages the Government has provided no specific evidence. The applicants have demonstrated that a number of pages are already in the public realm. It is to be remembered that section 48 of the Act places the burden of justifying confidentiality on the Government. For all of the reasons previously indicated, there is no clear, specific and understandable linkage between allegations of harm and the disclosure of these 323 specific pages. The Government has not, in my opinion, demonstrated reasonable grounds on which to refuse disclosure of these pages.
I should note that with respect to these 323 pages, the Information Commissioner recommended that 69[9] pages should not be disclosed in whole or in part. I have reviewed these pages but as I indicated earlier, the Court does not have the expertise of the Information Commissioner and in the absence of evidence, has no basis upon which to conclude that disclosure could reasonably be expected to cause harm to the Government. While the Court should have regard to the findings of the Information Commissioner, this does not equate to blind acceptance. The findings of the Information Commissioner must be brought to the Court in evidence by the party intending to rely upon them so as to enable the Court to understand why the Information Commissioner reached his conclusion in respect of specific pages. This has not been done in this case.
H. CONCLUSION
1. The approach of the Government witnesses indicates a well-intentioned attempt to avoid risk rather than to assess a reasonable expectation of harm from disclosure.
2. There is little in the way of clear and specific evidence linking a reasonable expectation of harm to the contents of specific pages.
3. There is only general reference or no express reference by the Government witnesses to approximately 90% of the record. This is insufficient to enable the Court to conclude there were reasonable grounds for confidentiality in respect of this information.
4. There is no specific assessment of overlaps between public and Government information. This is a relevant consideration. A comparison by the Court of Government information specifically referred to and public information reveals a high degree of overlap. The rationale advanced that disclosure of otherwise public information by the Government involves special risks is not persuasive.
I conclude that the head of the government institution did not have reasonable grounds to withhold disclosure of the record sought in this case. Each of the four applications is granted with costs.
[1] The applicants named as respondents the Prime Minister of Canada, the Privy Council Office and the Clerk of the Privy Council. I shall refer to these respondents collectively as the “Government”.
[2] One hundred and eight pages to all requesters and an additional twelve pages in respect of Mr. Rubin's request.
[3] On June 18, 1992 a further 35 pages were disclosed with 554 pages remaining confidential
[4] I am clearly bound by the jurisprudence. However the facts of the case at bar invite consideration of whether the test which has evolved may be too stringent and unyielding and that in cases where defences to disclosure are based on those sections which incorporate wording such as “could reasonably be expected to be injurious to”, that the seriousness of the consequences that may flow from disclosure should be taken into account.
[5] Of the 56 pages specifically referred to, 3 had been previously disclosed
[6] Of the 182 pages generally referred to, 4 had been previously disclosed.
[7]* In these instances, although I was unable to find a single question in the material submitted by the applicants that was the same as or similar to the one in the confidential information, I was able to locate questions whose subject-matter as a group encompassed the subject-matter of the confidential question.
[8] I am not suggesting that an “incorporation by reference” approach may not be appropriate in some cases. Where, for example, it is established that there is repetition of the same information on more than one page, detailed evidence in respect of one page may be sufficient with only brief reference to the other repetitive pages
[9] The Information Commissioner recommended that seventy-four pages remain confidential. Of these five were referred to specifically in the evidence. None were referred to generally. Sixty-nine were not referred to at all.