COURT OF APPEAL FOR ONTARIO  
CITATION: Ontario (Attorney General) v. Ontario (Information and Privacy  
Commissioner), 2022 ONCA 74  
DATE: 20220126  
DOCKET: C69150  
Gillese, Lauwers and Sossin JJ.A.  
BETWEEN  
Attorney General for Ontario  
Applicant (Appellant)  
and  
Information and Privacy Commissioner and Canadian Broadcasting Corporation  
Respondents (Respondents)  
Judie Im and Nadia Laeeque, for the appellant  
Justin Safayeni and Spencer Bass, for the respondent Canadian Broadcasting  
Company  
William S. Challis, for the respondent Information and Privacy Commissioner  
Daniel Sheppard, for the interveners Centre for Free Expression, Canadian  
Journalists for Free Expression, The Canadian Association of Journalists,  
Aboriginal Peoples Television Network, and News Media Canada  
Heard: August 31, 2021 by video conference  
On appeal from the order of the Divisional Court of Justices Swinton, Penny and  
Kristjanson, dated August 27, 2020, with reasons reported at 2020 ONSC 5085,  
affirming the decision of the Information and Privacy Commissioner, dated on  
July 15, 2019, with reasons reported at PA18-390.  
Page: 2  
Sossin J.A.:  
A. OVERVIEW  
[1]  
At issue in this appeal is whether the public is entitled to access the mandate  
letters provided to Cabinet ministers by the Premier of Ontario following the  
formation of the new government after the 2018 provincial election.  
[2]  
A journalist with the CBC requested access to the 23 letters sent by the  
newly elected Premier, Doug Ford, to Ontario’s Cabinet ministers who, together  
with him, comprise the Executive Committee, commonly known as the Cabinet.  
[3]  
Cabinet Office refused the CBC’s request. The CBC appealed the refusal to  
the Information and Privacy Commissioner of Ontario (the “IPC”). Mediation did  
not resolve the issues and so the parties proceeded to the adjudication stage,  
where they engaged in a lengthy process of written submissions.  
[4]  
The Cabinet Office opposed disclosure of the letters on the basis of the  
Cabinet privilege exemption under s. 12(1) of the Freedom of Information and  
Protection of Privacy Act, R.S.O. 1990, c. F.31. By Order PO3973; Cabinet Office  
(Re), [2019] O.I.P.C. No. 155, the Commissioner, Brian Beamish, ordered the  
letters to be disclosed to the requester, the CBC.  
[5]  
The Divisional Court dismissed the Attorney General for Ontario’s  
application for judicial review of the IPC Order, who now appeals to this court.  
Page: 3  
[6]  
Aside from the AG Ontario and CBC, this court heard submissions from the  
IPC on issues relating to the standard of review, the IPC process and the principles  
governing the interpretation of the Act. This court also heard submissions by a  
coalition of media and free-expression organizations as interveners. They  
appeared to help illustrate how the IPC’s approach protects cabinet deliberations  
and upholds the public right of access. The interveners Centre for Free Expression,  
Canadian Journalists for Free Expression, The Canadian Association of  
Journalists, and Aboriginal Peoples Television Network presented submissions  
before the Divisional Court and were granted leave to intervene before this Court,  
with the addition of News Media Canada.  
[7]  
For the reasons that follow, I would dismiss the appeal.  
The Legislation  
[8]  
Section 12(1) of the Act is particularly at issue in this appeal. It provides:  
12(1) A head shall refuse to disclose a record where the disclosure  
would reveal the substance of deliberations of the Executive Council  
or its committees, including:  
a. an agenda, minute or other record of the deliberations or  
decisions of the Executive Council or its committees;  
b. a record containing policy options or recommendations  
submitted, or prepared for submission, to the Executive  
Council or its committees;  
c. a record that does not contain policy options or  
recommendations referred to in clause (b) and that does  
contain background explanations or analyses of  
Page: 4  
problems submitted, or prepared for submission, to the  
Executive Council or its committees for their  
consideration in making decisions, before those  
decisions are made and implemented;  
d. a record used for or reflecting consultation among  
ministers of the Crown on matters relating to the making  
of government decisions or the formulation of  
government policy;  
e. a record prepared to brief a minister of the Crown in  
relation to matters that are before or are proposed to be  
brought before the Executive Council or its committees,  
or are the subject of consultations among ministers  
relating to government decisions or the formulation of  
government policy; and  
f. draft legislation or regulations.  
The Procedural History  
Cabinet Office provided the IPC with a copy of the Cabinet agenda that  
[9]  
indicated the Letters were distributed on July 11, 2018.  
[10] In detailed reasons, the IPC determined that the Letters did not fall within  
the exclusion set out in s. 12(1) of the Act. With respect to the content of the  
Letters, the IPC found:  
[79] The mandate letters are directives from the Premier  
to each of his ministers. They contain general statements  
about the government’s overall priorities and provide  
guidance to each minister as to each ministry’s priorities  
and his or her own role.  
[11] The IPC stated that the opening words of s. 12(1) required Cabinet Office to  
provide sufficient evidence to establish a link between Cabinet deliberations and  
Page: 5  
the content of the mandate letters, which are the records in issue. That is, if a  
record does not appear in the list from sub-paras. (a) to (f), s. 12(1) applies only if  
the record permits accurate inferences regarding actual Cabinet deliberations. It is  
not enough that the documents reveal the outcome of those deliberations; they  
must reveal its substance.  
[12] The IPC found no evidence that the records were tabled at a Cabinet  
meeting or that their contents were the subject of Cabinet deliberations. Even the  
agenda provided by the Cabinet Office did not indicate that the Letters were tabled  
for discussion. The Letters did not assess the reasons for or against a particular  
course of action, nor did they outline the views, opinions, thoughts, or ideas of  
cabinet members. Consequently, the Cabinet Office failed to show that the Letters  
would reveal the substance of any Cabinet deliberations.  
[13] The Divisional Court found the IPC’s decision to be reasonable. It  
characterized the case as an application of well-settled principles to particular  
facts. In the court’s view, the IPC used the absence of deliberation about the letters  
as evidence that s. 12(1) did not apply, not as a means to narrow the scope of the  
exclusion. Since the Letters did not disclose or invite any deliberation from Cabinet,  
the IPC’s conclusion that they did not meet the exception in the opening words of  
s. 12(1) was reasonable. Further, the Divisional Court found that the IPC did not  
apply an unreasonably stringent test by requiring Cabinet Office to show that the  
Page: 6  
Letters would be placed before specific Cabinet meetings in the future. The IPC  
simply found that the Cabinet Office had not discharged its burden to prove a link  
between the Letters and the substance of future Cabinet deliberations.  
B. DETAILED BACKGROUND  
The Media Request  
[14] The CBC made an access request to Cabinet Office under the Act for a copy  
of each of the Letters sent by the premier to Cabinet Ministers for all of Ontario’s  
22 ministries, and two non-portfolio responsibilities.  
The Cabinet Office Response  
[15] Cabinet Office denied the CBC access in full to the Letters, claiming the  
application of the mandatory exemption in s. 12(1) of the Act (Cabinet records).  
[16] Cabinet Office described the Letters in the following terms:  
Mandate letters are customarily the first communication  
to ministers through which the Premier translates party  
values and policy priorities into a plan of action for the  
government. For this reason, mandate letters outline the  
key policy priorities of the Premier that each minister is  
responsible for leading. Policy priorities are assigned to  
each minister based on the operational and/or statutory  
mandate of their ministry.  
In addition, mandate letters can include advice,  
instructions and guidance to each minister in carrying out  
his or her ministerial duties and responsibilities. This  
guidance is often placed in the context of the values that  
are important to the Premier and party.  
Page: 7  
Each member of the Executive Council who receives a  
mandate letter is accountable to the Premier and his or  
her Cabinet colleagues for assisting the government to  
achieve the priorities and objectives described in that  
letter.  
[17] Cabinet Office asserted that s. 12(1) applies where records reflect the  
policy-making and priority setting functions” of the Premier.  
[18] Cabinet Office submitted that the Letters were distributed to the Ministers at  
a July 11, 2018, meeting and took the position that disclosure of the Letters would  
reveal deliberations that took place in relation to the Letters at that Cabinet  
meeting.  
[19] Cabinet Office also submitted that the Letters included “the substance of  
deliberations of Cabinet” because the deliberations of the Premier, when setting  
policy priorities for Cabinet, are inherently part of Cabinet’s deliberative process.  
Cabinet Office further submitted that the Letters are exempt from disclosure  
because they would reveal the substance of future deliberations of Cabinet.  
The IPC Decision  
[20] The IPC rejected Cabinet Office’s argument that the Letters were exempt  
because they reflected the Premier’s policymaking and priority setting functions.  
[21] The IPC reviewed evidence regarding the preparation of previous mandate  
letters made publicly available in Ontario and elsewhere in Canada, and  
concluded, “[t]his evidence contradicts the view that disclosure of letters of this  
Page: 8  
nature would impinge on Cabinet deliberations; and I find no material difference in  
the nature of letters in issue before me to suggest a different result in this case.”  
[22] The IPC found no evidence to support Cabinet Office’s position that the  
Letters would reveal prior deliberations by the Premier or Cabinet Ministers, that  
they were discussed at a meeting of Cabinet, or that the Letters would be the basis  
of discussion at future Cabinet meetings.  
[23] After reviewing the submissions by the parties in detail, the IPC made the  
following findings of fact:  
a. There is nothing on the face of the Letters or in the representations of  
Cabinet Office to indicate that the Letters themselves were intended to  
serve, or did serve, as Cabinet submissions or as the basis for  
discussions by Cabinet as a whole: at para. 113.  
b. Cabinet Office provided no evidence that the Letters were themselves,  
in fact, discussed at the Cabinet meeting when they were provided to  
each minister or that they were tabled or made generally available for  
discussion: at para. 114.  
c. There is no evidence that the Letters were distributed to Cabinet as a  
whole at that time or that any specific contents of the Letters were  
actually the subject of the deliberations of Cabinet: at para. 114.  
d. The Letters do not reveal any discussions weighing or examining the  
reasons for or against a course of action with a view to making a  
decision: at para. 115.  
e. The Letters do not reveal any views, opinions, thoughts, ideas and  
concerns expressed by Cabinet members in the course of the  
deliberative process: at para. 115.  
Page: 9  
f. The Letters do not provide insight into the substance of any specific  
deliberations that may have occurred among Cabinet ministers: at  
para. 115.  
g. There is no evidence that the Letters themselves would be placed before  
Cabinet in future meetings: at para. 116.  
h. The Letters do not reveal the substance of any material upon which  
Cabinet members will actually deliberate in the future and do not reveal  
the substance of any such future deliberations: at para. 119.  
i. There is no persuasive evidence that disclosure would give rise to a  
chilling effect on Cabinet deliberations: at para. 123.  
j. The Letters do not reveal the substance of the Premier’s deliberations  
but, rather, the product of his deliberations: at para. 134.  
[24] Based on these findings of fact, the IPC held that s. 12(1) did not apply to  
the records at issue, and ordered Cabinet Office to disclose the letters to the  
appellant.  
The Divisional Court Decision  
[25] Assessing the IPC decision on a reasonableness standard, the Divisional  
Court dismissed the application for judicial review brought by the AG Ontario.  
[26] Writing for the Divisional Court, Penny J. explained that the Decision was  
largely fact-based, and resulted from the absence of evidence led by the  
AG Ontario that the Letters fell within the terms of s. 12(1) of the Act:  
[24] I cannot agree with the Attorney General that there  
is any fundamental error in the interpretation of the Act.  
In my view this is entirely a case of the application of well-  
settled principles to the particular facts. The burden of  
proof was undeniably on the government to demonstrate  
that the Letters fell within the s. 12(1) exemption. The  
Page: 10  
government chose to enter as evidence only the Letters  
themselves and a heavily redacted copy of the agenda  
for the meeting at which the Letters were, apparently,  
delivered. The IPC simply held that, on this record, the  
government had failed to satisfy its evidentiary burden.  
This is a sufficiency of evidence case, nothing more.  
[27] Throughout its decision, the Divisional Court relied on the IPC’s findings on  
the record to determine that the IPC Decision was reasonable. For example, with  
respect to whether the Letters were discussed at a Cabinet meeting, the Divisional  
Court held:  
[37] As is often said, inferences must be grounded in  
evidence from which the suggested inference may  
reasonably be drawn. Inferences unsupported by  
evidence, or which do not reasonably follow from the  
established facts, are mere speculation.  
[38] Regarding the meeting agenda itself, the subject of  
the Letters does not appear in the numbered list of  
agenda items (there are seven, all completely redacted).  
Rather, reference to the Letters appears at the end of the  
agenda under a heading “Chair Notes: Mandate Letters”.  
[39] Nothing about the content of this Note supports an  
inference that the mandate Letters were discussed at the  
Cabinet meeting. Rather, if anything, the content of the  
Note suggests the opposite.  
[40] As noted earlier, the Letters themselves do not  
suggest they are drafts subject to negotiation or in any  
way invite dialogue about their content. While it may be  
true that some of the mandates identified would likely  
require a return to Cabinet at some future time, this is  
nowhere specified or contemplated.  
[41] In these circumstances, there was a clear evidentiary  
basis to reject the Attorney General’s argument that it  
Page: 11  
was “reasonable to expect” certain unspecified aspects  
of the Letters “would have” been discussed at the initial  
Cabinet meeting. The IPC’s decision to do so was not  
unreasonable.  
[28] With respect to whether the Letters would reveal deliberations of future  
Cabinet meetings, the Divisional Court held:  
[49] Again, the IPC based this conclusion on an  
assessment of the evidence. The IPC simply found that  
the Cabinet Office had not discharged its burden to prove  
a link between the Letters and the substance of future  
Cabinet decisions. Given the paucity of evidence  
provided by the Cabinet office, this was not an  
unreasonable conclusion.  
[29] Finally, the Divisional Court rejected the AG Ontario submission that it was  
unreasonable for the IPC to require “stringent” evidence from Cabinet Office to  
support the argument that the Letters fell within the scope of s. 12(1) of the Act.  
On this point, writing for the court, Penny J. stated:  
[55] There is no merit to the Attorney General’s argument  
on this issue. The IPC clearly recognized, and applied,  
the correct standard of proof – it was the government’s  
onus to demonstrate that it met the requirements to come  
within the s. 12(1) exemption on a balance of  
probabilities. The Attorney General’s submission  
amounts to no more than an invitation for this Court to re-  
weigh the evidence and overturn the findings of the IPC  
with which the Cabinet Office disagrees. The IPC  
identified the correct legal principles, applied them to the  
interpretation of the opening words of s. 12(1), reviewed  
the record and the submissions before him in light of that  
legal test and explained the basis for his decision in  
thorough and cogent reasons. There was nothing  
Page: 12  
unreasonable about the IPC’s approach to or conclusions  
on the standard of proof.  
C.  
ISSUES ON APPEAL  
[30] The AG Ontario raises the following three issues in this appeal:  
a. The IPC erred in exercising the statutory authority to grant a right of  
access that is inconsistent with the purposes of the Act and the Cabinet  
records exemption;  
b. The IPC erred in exercising the statutory authority to grant a right of  
access based on an erroneous interpretation of s. 12(1) of the Act; and  
c. The IPC erred in exercising the statutory authority to grant a right of  
access based on an erroneous injection of a balancing test into s. 12(1)  
of the Act.  
[31] Each issue is addressed below.  
D.  
ANALYSIS  
The Standard of Review  
[32] On an appeal from an order of the Divisional Court concerning an application  
for judicial review of an administrative decision, this court must determine whether  
the Divisional Court identified the appropriate standard of review and applied it  
properly. In order to make this determination, the court “steps into the shoes” of  
the Divisional Court: see e.g., Canadian Federation of Students v. Ontario  
(Colleges and Universities), 2021 ONCA 553, at para. 20; Longueepee v.  
́
́
University of Waterloo, 2020 ONCA 830, 153 O.R. (3d) 641, at paras. 47-48,  
Page: 13  
applying Agraira v. Canada (Minister of Public Safety and Emergency  
Preparedness), 2013 SCC 36, [2013] 2 S.C.R. 559, at paras. 45-46.  
[33] The AG Ontario submits that, in upholding the IPC’s interpretation of s.12(1)  
of the Act, the Divisional Court failed to conduct a sufficiently “robust”  
reasonableness review as required by the Supreme Court in Canada (Minister of  
Citizenship and Immigration) v. Vavilov, 2019 SCC 65, 441 D.L.R. (4th) 1.  
[34] I do not accept this submission.  
[35] The Divisional Court acknowledged Vavilov as its point of departure for  
reviewing the reasonableness of the IPC decision. In identifying reasonableness  
as the standard of review, the Divisional Court stated, at para. 17:  
Reasonableness is the appropriate standard of review in  
this case. The reasonableness review finds its starting  
point in judicial restraint and respects the distinct role of  
administrative decision-makers.  
[36] I see no error in the Divisional Court’s approach to the standard of review.  
The IPC’s exercise of statutory authority to grant a right of  
access was consistent with the purposes of the Act and the  
Cabinet records exemption  
[37] The IPC described his approach to the interpretation of s. 12(1) as protective  
of communications within Cabinet’s deliberative process that would reveal the  
substance of its formulation of government policies.  
Page: 14  
[38] The IPC emphasized, however, that without additional evidence or context  
showing how the Letters would reveal the substance of deliberations, the  
introductory words of s. 12(1) did not shield the policy choices themselves.  
[39] The Divisional Court approached the issue as to the proper interpretation of  
s. 12(1) as one on which the parties agreed:  
[19] It is also accepted by the parties and the IPC that in  
order for the exemption under s. 12(1) to apply,  
disclosure of the record must “reveal the substance of  
deliberations” of Cabinet or “permit the drawing of  
accurate inferences” about past or future Cabinet  
deliberations. It is also accepted that the use of the term  
“including” in the introductory words of s. 12(1) means  
that any record which would reveal the substance of  
deliberations or permit the drawing of accurate  
inferences qualifies for the exemption; the specifically  
enumerated categories of record in subparagraphs (a) to  
(f) must be interpreted as providing an expanded  
definition of, or at the very least the removal of any  
ambiguity about, the types of records that are exempt  
from disclosure.  
[40] The AG Ontario takes issue with this characterization. It argues that the IPC  
erred by taking a narrow and restrictive view of the opening words of s. 12(1)  
inconsistent with the purposes of the Act and of the Cabinet records exemption.  
[41] The AG Ontario argues that s. 12(1) should be read in concert with the  
specific examples of exclusions set out in the subparagraphs (a) through (f), and  
in particular “(a) an agenda, minute or other record of the deliberations or decisions  
of the Executive Council or its committees.”  
Page: 15  
[42] The AG Ontario submits that the use of the term “including” in the  
introductory portion of s. 12 makes clear that the records set out in the  
subparagraphs constitute records whose disclosure would reveal the substance  
of deliberations of the Executive Council or its committees.” It characterizes this  
interpretive approach as the “illustrative approach.” Therefore, s. 12(1) exempts  
the Letters from disclosure by analogy to the records described in subparagraphs  
(a) to (f).  
[43] In contrast, the CBC, in supporting the approach taken by the IPC as  
reasonable, characterized the proper approach to the interpretation of s. 12(1) as  
the “expansive approach.” Under the expansive approach, the use of the term  
“including” in s. 12(1) indicates that the records specified in the following  
subparagraphs expand on the general language in the introductory portion of  
s. 12(1) by setting out records which, while not necessarily revelatory of the  
deliberations of Cabinet or its committees, may be so in certain circumstances and  
are, therefore, exempt from disclosure. Consequently, CBC argued that the letters  
fall outside of s. 12(1)’s scope because they do not reveal the substance of Cabinet  
deliberations or meet any of the expanded categories set out in subparagraphs (a)  
to (f).  
[44] As both the CBC and IPC note in their submissions, the expansive approach  
is in keeping with a long line of decisions by the IPC dealing with exemptions under  
Page: 16  
s. 12(1). For example, in Order P-266, [1991] O.I.P.C. No. 10, at p. 7, the IPC  
explicitly stated that the “use of the word ‘including’ in subsection 12(1) of the Act  
[sic] should be interpreted as providing an expanded definition of the types of  
records which are deemed to qualify as subject of the Cabinet records exemption,  
regardless of whether they meet the definition found in the introductory wording of  
subsection 12(1)”: citing Order 22, [1988] O.I.P.C. No. 22 (emphasis added).  
[45] Indeed, the IPC has used the same words to describe s.12(1) since Order  
P-901, [1995] O.I.P.C. No. 148. In that decision, at p. 4, the IPC stated:  
[T]he use of the term "including" in the introductory  
wording of section 12(1) means that the disclosure of any  
record which would reveal the substance of deliberations  
of the Executive Council or its committees (not just the  
types of records listed in the various parts of  
section 12(1)), qualifies for exemption under section  
12(1).  
[46] Other orders have held that a record which has never been placed before  
an Executive Council or its committees may nonetheless qualify for exemption  
under the introductory wording of s. 12(1): see e.g. Interim Order PO-1742-I, [2000]  
O.I.P.C., at para. 36; Order PO-2707; Ministry of Education, [2008] O.I.P.C.  
No. 166, at para. 26. This result will occur where a government organization  
establishes that the disclosure of the record would reveal the substance of  
deliberations of an Executive Council or its committees, or that its release would  
Page: 17  
permit the drawing of accurate inferences with respect to the substance of  
deliberations of an Executive Council or its committees.  
[47] As the Divisional Court highlighted, the subparagraphs of s. 12(1) “clarify  
that the exemption applies to specific types of records that might otherwise be  
thought to fall outside the opening words”: at para. 27.  
[48] The IPC’s Order P-901 further stands for the proposition that s. 12(1) is not  
limited to its subparagraphs. Any record can fall under the Cabinet records  
exemption so long as it would reveal the substance of deliberation of an Executive  
Council or its committees, or permit the drawing of accurate inferences.  
[49] While previous IPC decisions do not bind the IPC in relation to future  
interpretations of s.12(1), the IPC’s consistency in its approach to its governing  
statute may be taken as an indicator of the reasonableness of this decision. As the  
Supreme Court stated in Vavilov:  
[129] Administrative decision makers are not bound by  
their previous decisions in the same sense that courts are  
bound by stare decisis. As this Court noted in Domtar, “a  
lack of unanimity is the price to pay for the  
decision-making freedom and independence” given to  
administrative decision makers, and the mere fact that  
some conflict exists among an administrative body’s  
decisions does not threaten the rule of law: p. 800.  
Nevertheless, administrative decision makers and  
reviewing courts alike must be concerned with the  
general consistency of administrative decisions. Those  
affected by administrative decisions are entitled to expect  
that like cases will generally be treated alike and that  
Page: 18  
outcomes will not depend merely on the identity of the  
individual decision maker expectations that do not  
evaporate simply because the parties are not before a  
judge. [Citations omitted.]  
[50] Even if this interpretation of s.12(1) were not long-standing, in my view, the  
IPC would be acting reasonably in adopting it. The use of the term “including” prior  
to setting out the ss. 12(1)(a) to (f) is ambiguous.  
[51] In the face of ambiguous wording, the obligation on an administrative  
decision-maker is to provide a reasoned explanation for the interpretation adopted  
that is alive to the text, context and purpose of the provision; Vavilov, at para. 120.  
The IPC has done this.  
[52] The IPC’s approach to s. 12(1) is also consistent with the general purpose  
of the Act, which is to “provide a right of access to information under the control of  
institutions in accordance with the principles that…necessary exemptions from the  
right of access should be limited and specific”. This is a point underscored by the  
interveners, who argue that provisions of the Act which exempt access should  
generally be interpreted narrowly.  
[53] Before moving to the IPC’s exercise of statutory authority, I will briefly  
address two of the AG Ontario’s arguments which were not put before the IPC.  
[54] First, the AG Ontario points to the French translation of the term “including”  
in the Act, which is “notamment,” as opposed to “en outre.” The AG Ontario argues  
Page: 19  
that, in other statutory settings, “en outre” is used to indicate an expansive rather  
than inclusive set of specified subsections.  
[55] As the AG Ontario could have, but did not make this argument before the  
IPC, it should not be determinative of a finding that the IPC acted unreasonably in  
its interpretation: see Alberta (Information and Privacy Commissioner) v. Alberta  
Teachers' Association, 2011 SCC 61, [2011] 3 S.C.R. 654, at paras. 22-26.  
[56] However, even if it had been argued earlier, the French version of the  
provision is not conclusive of a particular, proper interpretation of s. 12(1) of the  
Act. The French dictionary definitions of “notamment” and “en outre” both refer to  
“including”. Thus, neither rule in nor rule out either interpretive approach.  
[57] Second, the AG Ontario raised the legislative history of s. 12(1), and the fact  
that, in the debates surrounding draft legislation preceding the Act, the legislature  
rejected a proposal to use the term “solely” rather than “including” to limit the  
exemption to specific records. The AG Ontario argues that this aspect of the  
legislative history provides further support for the illustrative approach.  
[58] Again, this legislative history argument was not before the IPC. In any event,  
this argument also fails to establish that the IPC’s adoption of the expansive  
approach was unreasonable. While the legislative history establishes that the term  
“solely” was not adopted, it does not lead to the conclusion that the term “including”  
is capable of only one reasonable interpretation.  
Page: 20  
[59] In my view, the AG Ontario has failed to establish that the IPC’s  
interpretation of s. 12(1) was unreasonable, and that the Divisional Court erred in  
finding the IPC’s interpretation reasonable.  
The IPC’s exercise of its statutory authority under s. 12(1) of the  
Act was reasonable  
[60] The AG Ontario argues that the IPC’s decision to order disclosure of the  
Letters was unreasonable as the letters represented “decisions” of the Premier  
which would reveal the substance of the deliberations leading up to these  
decisions. In its factum, the AG Ontario describes the Letters in the following terms:  
[23] The Letters are a way by which this Premier has  
chosen to discharge his constitutional duties as first  
minister to develop and prioritize the policies and  
operational agenda of the new government. The Letters  
represent the first communication to the ministers from  
the Premier to establish policy priorities and a plan of  
action for their development and implementation over the  
term of the current government.  
[25] In addition to setting out the policy priorities of the  
new government, the Letters also include opinion,  
advice, instructions and guidance from the Premier to the  
newly appointed ministers as to how to carry out their  
new ministerial duties and responsibilities. Each member  
of the Executive Council who received a Letter is  
accountable to the Premier and their other Cabinet  
colleagues for assisting the government in achieving the  
policy objectives described in the Letters.  
Page: 21  
[61] The CBC takes the position that the AG Ontario has failed to lead any  
evidence that the disclosure of the Letters would reveal the substance of  
deliberations, either the Premier’s or Cabinet’s.  
[62] Beside the Letters themselves, the AG Ontario relies on the agenda for the  
July 11, 2018 Cabinet meeting, which refers to the distribution of the mandate  
letters as a “Chair note.”  
[63] As noted above, the AG Ontario argued before the IPC that the Letters meet  
the threshold for excluded records under s. 12(1) of the Act on three grounds:  
a. the Letters disclosed the deliberations of the Premier in setting Cabinet’s  
policy priorities which are inherently part of the deliberative process of  
Cabinet;  
b. the Letters were the subject of deliberations at the meeting of Cabinet  
where the letters were placed on the agenda; and  
c. the deliberations at future Cabinet meetings where the policy priorities  
set out in the letters would be further discussed.  
[64] With respect to AG Ontario’s second argument and the reference to Letters  
in a Cabinet agenda, the IPC’s interpretation of s. 12(1) precluded the argument  
that all matters on a Cabinet agenda were presumptively excluded. Such a finding  
might flow from the illustrative approach to the subsection s. 12(1)(a) to (f), and the  
reference specifically in s. 12(1)(a) to “an agenda, minute, or other record of the  
deliberations or decisions of the Executive Council or committees” (emphasis  
added). However, the IPC rejected this approach.  
Page: 22  
[65] Similarly, the IPC found no evidence to support AG Ontario’s first and third  
arguments. The IPC reiterated that the mere stating of a policy priority does not  
reveal the deliberations leading to that outcome.  
[66] The IPC found that the AG Ontario provided no evidence that the Letters  
were actually discussed at the Cabinet meeting where their distribution was  
referenced in the agenda. Further, the IPC found that there was no evidence  
offered for the assertion that the Letters were tabled for future discussion by  
Cabinet.  
[67] The IPC also found no basis for the argument that the Letters themselves  
revealed the deliberations of the Premier.  
[68] The AG Ontario argued before the IPC that the Premier’s articulation of  
policy priorities represented a step in Cabinet’s deliberative “continuumand  
should not be seen as the culmination of the government’s policy decision-making.  
According to the AG Ontario, the deliberative process of the Executive Council  
continues once the Premier establishes and communicates his policy priorities to  
ministers through the Letters. The policy priorities will involve further deliberation  
and decision-making by Cabinet in future.  
[69] The IPC accepted that, where a record discloses deliberations by the  
Premier, this record may be exempted under s. 12(1). In other words, IPC  
recognized the distinct role of the Premier in relation to Cabinet. But the IPC  
Page: 23  
rejected the AG Ontario’s view of the deliberative process as unduly broad. In its  
decision, the IPC stated:  
[121] The submission advanced by Cabinet Office that  
the mandate letters “open the dialogue” and initiate a  
continuum of the deliberative process as a “blueprint” for  
future Cabinet discussions suffers from the same  
deficiency. I am asked to accept that deliberations on  
“nearly all” of the policy initiatives would take place at  
some point in future Cabinet meetings. I am also asked  
to find that section 12(1) applies to policy initiatives that  
may never return to Cabinet at all or that may be altered  
or amended in significant and unspecified ways. With  
respect, Cabinet Office has it backwards. I must be  
satisfied on the evidence of the likelihood that that  
disclosure of the letters “would reveal” or, at a minimum,  
permit accurate inferences to be drawn concerning the  
substance of future Cabinet deliberations.  
[122] That is not to say that deliberations will not ensue a  
later date in relation to the subject matter of certain  
priorities. However, any such deliberations would be in  
relation to proposals or other materials yet to be  
developed by individual ministers and later brought  
before Cabinet. Such materials, when developed, may  
well reveal the substance of future Cabinet deliberations  
if and when they occur. However, the evidence before me  
does not establish that disclosure of the mandate letters  
themselves will permit accurate inferences to be drawn  
in that respect. At most, Cabinet Office’s submissions  
indicates that the subject matter of future deliberations  
may be revealed by disclosure.  
[70] On the question of the “continuum” approach to disclosure of deliberations,  
Penny J. stated:  
[29] As to the Attorney General’s “continuum” argument,  
the introductory words of s. 12(1) do not protect all  
Page: 24  
records leading up to any particular government decision;  
they protect the substance of deliberations of Cabinet  
(which includes, as found previously by the IPC, the  
Premier’s deliberations in setting Cabinet’s priorities).  
The Letters, on their face however, do not disclose or  
invite any deliberative process. The Cabinet Office’s own  
submissions describe the Letters as “the culmination of  
an extensive deliberative process by the Premier [that]  
reflect his/her determination, as first minister, of the  
priorities of the new government”. In the absence of any  
other evidence, the IPC’s conclusion that the Letters do  
not disclose deliberative processes was a reasonable  
one. [Emphasis in original.]  
[71] The IPC also considered its own previous decisions with respect to the role  
of the Premier’s Office. In Order PO-1725, [1999] O.I.P.C. No. 153, the IPC  
considered a request for a scheduling book prepared by a Premier Office senior  
staff member. After a lengthy review of the Premier’s role and the importance of  
his or her staff, the IPC concluded that the records fell under s. 12(1).  
[72] In this case, the issue for the IPC was not whether records which disclose  
the deliberations of the Premier are caught by the exclusion under s. 12(1), but  
rather whether the Letters and the agenda constitute sufficient evidence that the  
deliberations of the Premier would be revealed by disclosure of the Letters.  
[73] On this point, at para. 132, the IPC found on the record before him that the  
disclosure of the policy initiatives in the mandate letters would not provide any  
insight into the deliberative considerations or consultative process by which the  
Premier arrived at them.”  
Page: 25  
[74] For the Divisional Court, Penny J. distinguished the IPC Order PO-1725 as  
well, stating, at para. 31:  
The decision of the IPC in Order PO 1725 does not  
support the Attorney General’s argument. Again, this is  
on essentially factual and evidentiary grounds. In Order  
PO 1725, the IPC found that the Premier’s  
“consultations with a view to establishing Cabinet  
priorities are an integral part of Cabinet’s substantive  
deliberative process” and that the records reflecting  
those “consultations” constitute the “substance of  
deliberations”. It was this deliberative or consultative  
aspect of the Premier’s priority-setting process which lay  
at the heart of the IPC’s decision in that case. There is no  
evidence of any such consultative or deliberative process  
in establishing the Premier’s priorities here. In fact, in  
Order PO 1725, the IPC specifically found that (apart  
from the formal agenda document itself) the subject  
matter of items considered or to be considered by  
Cabinet will not “normally be found to reveal the  
substance of Cabinet deliberations, unless either the  
context or other additional information would permit the  
reader to draw accurate inferences” as to actual  
deliberations which took place at a particular Cabinet  
meeting. [Emphasis added.]  
[75] I agree with the IPC and the Divisional Court. The scheduling book at issue  
in Order PO-1725 was far closer to the Premier’s deliberative process than the  
Letters at issue here. The scheduling book contained “references to particular Bills  
or pending legislation, [and] more generalized references to possible programs  
and initiatives”: at p. 15. Therefore, it is apparent that the IPC applied s. 12(1) in  
Order PO-1725 because the scheduling book revealed the thoughts and opinion  
of the Premier and, consequently, Cabinet.  
Page: 26  
[76] Conversely, the Letters are the culmination of that deliberative process.  
While they highlight the decisions the Premier ultimately made, they do not shed  
light on the process used to make those decisions, or the alternatives rejected  
along the way. Accordingly, the Letters do not threaten to divulge Cabinet’s  
deliberative process or its formulation of policies.  
[77] In my view, this application of s. 12(1) by the IPC to the Letters was  
reasonable, and the Divisional Court committed no error in so finding.  
The IPC did not introduce a new balancing test into the exercise  
of its statutory authority under s. 12(1) of the Act  
[78] The AG Ontario argued that the IPC injected a balancing test into s. 12(1)  
despite its exclusion from the “public interest override.”  
[79] The AG Ontario refers to s. 23 of the Act which permits disclosure of exempt  
records if there is “a compelling public interest in disclosure that clearly outweighs  
the purpose of the exemption.” Section 23, however, expressly does not apply to  
Cabinet records that are otherwise caught by s. 12.  
[80] The AG Ontario argues that the IPC’s reference to the “public interestas a  
balancing factor, relying in part on the Nova Scotia Court of Appeal in O’Connor v.  
Nova Scotia, 2001 NSCA 132, 197 N.S.R. (2d) 154, thus constitutes a reversible  
error.  
[81] I disagree.  
Page: 27  
[82] In my view, the IPC did not inject a new balancing test into the analysis of  
s. 12(1). Rather, the IPC, relying on O’Connor, recognized that s. 12(1) itself  
strikes a balance between a citizen’s right to know what government is doing and  
a government’s right to consider what it might do behind closed doors: Order PO-  
3973, at para. 97. The Court, at para. 1 in O’Connor, stated that this context calls  
for an interpretation of the Act that attempts to balance these two public rights.  
[83] While the Divisional Court highlighted the factual distinctions between this  
case and O’Connor, where there was substantial evidence that the records at issue  
would reveal Cabinet deliberations, the reference to the balance reflected in the  
Act is appropriate in the context of this case as well.  
[84] I see no error in the IPCs reference to these general observations by the  
Nova Scotia Court of Appeal in O’Connor as part of its analysis. His reference  
focuses on the Court of Appeal’s analysis linking the records at issue with the  
substance of deliberation. That is exactly the thrust of the opening words exception  
in s. 12(1).  
[85] Generally, the AG Ontario submits that the Divisional Court erred in failing  
to conduct a proper Vavilov review of the IPC’s statutory analysis.  
[86] I would not accept this submission.  
[87] The Divisional Court committed no error arising from Vavilov in its finding  
that the Decision was reasonable.  
Page: 28  
DISPOSITION  
[88] For the reasons set out above, I would dismiss the appeal.  
[89] If the AG Ontario and CBC cannot agree on costs, brief written submissions  
may be provided to the court (not to exceed three pages double-spaced) within  
15 days of the release date of these reasons.  
[90] Neither the interveners nor the IPC sought costs and I would order none.  
“L. Sossin J.A.”  
“I agree. E.E. Gillese J.A.”  
Page: 29  
Lauwers J.A. (dissenting):  
A.  
OVERVIEW  
[91] The Information and Privacy Commissioner ordered Cabinet Office to  
disclose to the CBC the mandate letters Premier Ford gave to Cabinet ministers  
at a Cabinet meeting. The distribution of the letters was on the meeting’s agenda.  
The Commissioner determined that the exemption from disclosure for Cabinet  
records set out in the Freedom of Information and Protection of Privacy Act1 did  
not apply to the mandate letters. The issue is whether the Commissioner’s  
interpretation of the exemption was reasonable.  
[92] Section 12(1) of the Act sets out the relevant exemption from public  
disclosure for Cabinet records: “A head shall refuse to disclose a record where the  
disclosure would reveal the substance of deliberations of the Executive Council or  
its committees, including…”. These opening words are followed by subparagraphs  
(a) to (f), which list specific kinds of records that are exempted, such as agendas  
or minutes of Cabinet deliberations.  
[93] The purpose of the exemption is to establish a robust and well-protected  
sphere of confidentiality within which Cabinet can function effectively, one that is  
consistent with the established conventions and traditions of Cabinet government.  
1 Freedom of Information and Protection of Privacy Act, R.S.O. 1990, c. F.31, s. 12(1).  
Page: 30  
I conclude that the Commissioner’s interpretation was unreasonable because the  
effect of his order, contrary to the legislature’s intention, was to breach, erode, or  
undermine those traditions. I therefore dissent.  
[94] Because the mandate letters are part of the Premier’s deliberative process,  
in his role as head of Cabinet, and initiate Cabinet’s mandate going forward, they  
are, perforce, part of Cabinet’s deliberative process. They are records that would  
reveal the nature of Cabinet deliberations understood as including the topics,  
subject matters, things, or “body of information” Cabinet would be discussing, as  
well as the deliberations themselves within the meaning of “the substance of  
deliberations” in the opening words of s. 12(1). This is dispositive of the appeal,  
which I would allow on this basis alone.  
[95] I would also allow the appeal on the grounds that the Commissioner  
prescribes a new test that a record must meet to qualify for exemption from  
disclosure under the opening words of s. 12(1): “If a record does not appear at  
paragraphs (a) to (f), it will only qualify for the exemption if the context or other  
information would permit accurate inferences to be drawn as to actual Cabinet  
deliberations at a specific Cabinet meeting.”2 This new test is fundamentally  
incompatible with the text, context, and purpose of s. 12(1) of the Act.  
2 Order PO-3973; Cabinet Office (Re), [2019] O.I.P.C. No. 155, at para. 101 (emphasis added).  
Page: 31  
[96] Turning to my colleague’s reasons, he states that “[t]he use of the term  
‘including’ prior to setting out the ss. 12(1)(a) to (f) is ambiguous.”3 He takes the  
position that this ambiguity opens up a policy choice that the Commissioner is  
entitled to make under the Act between two competing approaches, one broad and  
one narrow, and the Commissioner did not err in preferring the narrow approach  
to limit the exemption.  
[97] I disagree. The ostensible presence of an ambiguity in s. 12(1) of the Act  
does not open up a policy choice for the Commissioner to make. The legislature  
made the policy choice as to the reach of the protected sphere of Cabinet  
confidentiality in enacting s. 12(1). The Commissioner’s task was to identify and  
apply that legislative choice. This he failed to do, as did the Divisional Court.  
B.  
THE ISSUES  
[98] I frame my analysis around this sequence of questions:  
1) What are the relevant principles of statutory interpretation and what  
is the applicable standard of review?  
2) What is the pertinent context within which s. 12(1) of the Act is to be  
interpreted?  
3) What is the purpose of s. 12(1)?  
4) In light of that context and that purpose, what interpretation of the  
text of s. 12(1) should the Commissioner have adopted?  
5) Is the Commissioner’s new test compatible with s. 12(1)?  
6) Should the disclosure of the mandate letters be remitted to the  
Commissioner for disposition in accordance with these reasons?  
3 Reasons of Sossin J.A., at para. 50.  
Page: 32  
C.  
THE ANALYSIS  
[99] I begin with two observations. First, Cabinet mandate letters are a relatively  
recent form of political document issued by prime ministers and premiers. They  
have been used as a form of public letter to frame an area of public policy and  
often reiterate campaign promises. Of course, prime ministers and premiers are  
free to craft and issue such letters for public consumption. But the issue in this  
case is different. Can the Commissioner compel the disclosure of mandate letters,  
in the face of the exemption from disclosure for Cabinet records in s. 12(1) of the  
Act, when the Premier chooses to keep them private? This issue engages  
constitutional conventions and traditions surrounding the Premier’s role in matters  
pertaining to Cabinet deliberations, and it has broader implications.  
[100] My second observation is that an ultimate decision forcing disclosure of the  
mandate letters in this case is likely to be a one-off. Why do I say that? Because  
the Premier’s response in the future will predictably take one of three forms: to  
draft mandate letters for purely public consumption as others have done; to tie  
mandate letters even more closely to the Cabinet decision-making process in order  
to better substantiate the claim to an exemption from public disclosure under  
s. 12(1) of the Act; or to give up drafting mandate letters altogether.  
[101] That said, the Commissioner’s incursion into the ordinary operations of  
Cabinet is not benign or unimportant, and it should not be trivialized. The  
Page: 33  
Commissioner overstepped and the Divisional Court was wrong to uphold his  
decision. It was neither the legislature’s intention nor the purpose of s. 12(1) of the  
Act to force Cabinet to change its customary way of operating. The basis on which  
the Commissioner overstepped could give rise to future problems of a markedly  
more serious nature than the disclosure of mandate letters might suggest.  
[102] I now turn to the questions that frame my analysis.  
What are the relevant principles of statutory interpretation and  
what is the applicable standard of review?  
[103] The interpreter’s task in statutory interpretation is to discern the legislature’s  
intention in order to give effect to it.4 The interpreter must attend to text, context,  
and purpose.5  
[104] Section 1 of the Act stipulates two purposes. The first purpose is: “(a) to  
provide a right of access to information under the control of institutions in  
accordance with the principles that, (i) information should be available to the public,  
[and] (ii) necessary exemptions from the right of access should be limited and  
specific”. The second purpose to protect the privacy of individuals is not  
engaged in this appeal.  
4 Canada (Minister of Citizenship and Immigration) v. Vavilov, 2019 SCC 65, 441 D.L.R. (4th) 1,  
at paras. 117, 121.  
5 Vavilov, at paras. 118-24.  
Page: 34  
[105] Section 12(1) of the Act sets out the relevant exemption from public  
disclosure for Cabinet records. The interpreter must reconcile the public access to  
information purpose of the Act set out in s. 1(a) with the purpose for the exemption  
from disclosure for Cabinet records set out in s. 12(1).  
[106] The standard of review to be applied to a specialized tribunal’s statutory  
interpretation is ordinarily reasonableness.6 However, because questions  
regarding “the relationship between the legislature and the other branches of the  
state… require a final and determinate answer from the courts,7 when the  
interpretation engages a constitutional question, the standard is correctness.  
[107] This case presents a conundrum. Constitutional conventions are engaged,  
which gives a constitutional dimension to the interpretation exercise. But  
constitutional conventions are not law beyond a legislature’s reach. Good  
constitutional order requires at least a presumption that the legislature did not  
intend to abrogate any constitutional conventions absent a clear signal to the  
contrary. All the signals in the Act’s development and in its text are in the direction  
of respect for those conventions and the associated traditions and practices.  
[108] I conclude that the Commissioner’s interpretation of s. 12(1) was  
unreasonable. I would leave for another day the thorny question of whether the  
6 Vavilov, at paras. 115, 119.  
7 Vavilov, at para. 55.  
Page: 35  
constitutional overlay in this case that is, the constitutional conventions and  
associated traditions and practices surrounding the role of the Premier in matters  
pertaining to Cabinet deliberations requires the Commissioner to be correct in  
his interpretation.  
What is the pertinent context within which s. 12(1) of the Act is  
to be interpreted?  
[109] The word “context” in the phrase, “text, context, and purpose”, has an  
external dimension, outside the Act’s text, which positions the legislation in the  
larger world. The context also has an internal dimension by which the Act as a  
whole must be given a coherent interpretation that reconciles its access to  
information purpose in s. 1(a) with the purpose for the Cabinet records disclosure  
exemption in s. 12(1). I begin with the external dimension and then turn to the  
internal.  
(a) The external contextual dimension  
[110] In this section of the reasons, I consider the constitutional context and then  
the policies supporting public access to government information.  
(i)  
The constitutional context  
[111] The policy work behind the Act was largely done by the Commission on  
Freedom of Information and Individual Privacy, which was headed by Dr. D.  
Page: 36  
Carlton Williams.8 The Williams Report noted that the “terms of reference directed  
us to consider possible changes in public information practices which would be  
‘compatible with the parliamentary traditions of the Government of Ontario.’”9  
These include those constitutional conventions and traditions surrounding the role  
of the Premier in matters pertaining to Cabinet deliberations. Such conventions  
and traditions form the deep contextual backdrop, which must not be ignored,  
forgotten, or paid mere lip service. The Report was sensitive to the political realities  
of Cabinet government and to how access to information should function, as is the  
Act, properly interpreted. The Commissioner was not similarly sensitive, nor was  
the Divisional Court.  
The Westminster model of responsible government in Canada  
[112] The Constitution Act, 1867 established a modified Westminster model of  
responsible government in Canada via the preamble, which mandates “a  
Constitution similar in Principle to that of the United Kingdom.” As Professor Peter  
Hogg notes: “[T]he rules which govern [responsible government in Canada] are  
almost entirely ‘conventional’, that is to say, they are not to be found in the ordinary  
legal sources of statute or decided cases.10 The various elements of the  
8 Public Government for Private People: The Report of the Commission on Freedom of Information and  
Individual Privacy, vol. 2 (Toronto: Queen’s Printer of Ontario, 1980) (the “Williams Report” by the  
“Williams Commission”).  
9 Williams Report, at p. 83.  
10 Peter W. Hogg, Constitutional Law in Canada, loose-leaf, 5th ed. (Toronto: Thomson Reuters Canada  
Ltd., 2007), at para. 9-3. The most significant modification is the country’s federal structure. In this  
Page: 37  
Westminster model form a tradition. A tradition is like an iceberg. The bulk of it is  
not immediately visible. Understanding a tradition in order to do no harm to it  
requires careful, attentive, and sensitive work.  
[113] I accept Hogg’s general description of the Canadian version of the  
Westminster model of responsible government. He observes that “the forms of  
monarchical government are retained, but real power is exercised by the elected  
politicians who give advice to the Queen and her representatives.11 By  
convention, the Governor General selects as Prime Minister the “person who can  
form a government that will enjoy the confidence of the House of Commons.12  
Hogg notes: “Responsible government transfers the real power to the elected  
Prime Minister”, who is the “political head of state”.13  
[114] The Prime Minister has two significant powers. The first is “the power to  
select the other ministers, and the power to promote, demote or dismiss them at  
pleasure.14 The Governor General appoints the ministers on the Prime Minister’s  
section, I pick out of Professor Hogg’s text some pertinent descriptive statements that are indisputable,  
though not, as he points out, absolute or without exception. The concepts applicable to the federal  
government apply with necessary modifications to the provinces; the Premiers are the Prime Minister’s  
equivalent: Hogg, at paras. 9-1, 9-3. Ontario’s Executive Council, the provincial equivalent of the Privy  
Council, is mandated by the Executive Council Act, R.S.O. 1990, c. E.25, although a number of other  
pieces of legislation affect its composition and functions: see F.F. Schindeler, Responsible Government in  
Ontario (Toronto: University of Toronto Press, 1969), at p. 30.  
11 Hogg, at para. 9-1.  
12 Hogg, at para. 9-4.  
13 Hogg, at para. 9-1.  
14 Hogg, at para. 9-6.  
Page: 38  
advice.15 The second power is to seek dissolution for an election.16 Hogg observes  
that these powers, along with the “special authority” that comes from having won  
an election, “ensures that the Prime Minister’s voice will be the most influential one  
within the cabinet.17  
[115] The appointed ministers meeting together as a group constitute Cabinet,  
which is “in most matters the supreme executive authority.”18 Functionally, Cabinet  
“formulates and carries out all executive policies, and it is responsible for the  
administration of all the departments of government.”19 Hogg adds that full  
Cabinet’s role in decision-making “may depend in large measure upon the  
discretion of the Prime Minister” because “the Prime Minister calls the meetings of  
cabinet, settles the agenda, presides over the meetings, and ‘defines the  
consensus’ on each topic.20 Accordingly, “[t]he Prime Minister (or provincial  
Premier) effectively controls the executive branch of government through his  
control over ministerial appointments and over the cabinet.21  
15 Hogg, at para. 9-4.  
16 Hogg, at para. 9-6.  
17 Hogg, at para. 9-6.  
18 Hogg, at para. 9-5.  
19 Hogg, at para. 9-5.  
20 Hogg, at para. 9-5 (footnote omitted).  
21 Hogg, at para. 9-6.  
Page: 39  
The separation of powers  
[116] The constitutional doctrine of the separation of powers has applied since  
Confederation. The courts have policed the division of powers and, since 1982,  
also compliance with the Charter.22 Although the separation of powers in Canada  
is not strict, Canadian constitutional law “recognize[s] and sustain[s] some notion  
of the separation of powers.23  
[117] The three branches are the executive, the legislative, and the judicial.24 Most  
of the case law on the separation of powers has considered the line between the  
judicial and the legislative branches, which is necessary to ensure impartial justice.  
The line between the executive and the legislative is less distinct and has been  
addressed less often.  
[118] Karakatsanis J. observed: “All three branches have distinct institutional  
capacities and play critical and complementary roles in our constitutional  
democracy.”25 She added a relevant caution: “However, each branch will be unable  
to fulfill its role if it is unduly interfered with by the others.” Karakatsanis J. cited the  
words of McLachlin J.: “It is fundamental to the working of government as a whole  
22 Ontario v. Criminal Lawyers’ Association of Ontario, 2013 SCC 43, [2013] 3 S.C.R. 43, at paras. 27-31,  
a decision that concerned the appointment of amicus curiae by judges.  
23 Cooper v. Canada (Human Rights Commission), [1996] 3 S.C.R. 854, at paras. 10-11, per Lamer C.J.  
And see Côté J.’s partially dissenting reasons in References re Greenhouse Gas Pollution Pricing Act,  
2021 SCC 11, 455 D.L.R. (4th) 1, at para. 279.  
24 The precise status in the constitutional pantheon of certain officials created by statute, such as the  
Auditor General, the Ombudsman and the Information and Privacy Commissioner, is unclear.  
25 Criminal Lawyers’ Association of Ontario, at para. 29.  
Page: 40  
that all these parts play their proper role”, to which McLachlin J. added her own  
caution: “It is equally fundamental that no one of them overstep its bounds, that  
each show proper deference for the legitimate sphere of activity of the other.”26  
[119] A functional and purposive approach must be taken to the question of what  
powers and privileges each branch has in relation to the others. The analysis must  
engage what each branch functionally needs in order to perform its expected role  
within the constitutional polity. This flows ineluctably from the nature of the  
Westminster model of responsible government, as the cases have recognized.  
The executive-legislative separation  
[120] Some commentators, including Hogg, posit that in the Westminster model,  
“there is no separation of powers between the executive and legislative branches”  
because “[t]he head of the executive branch, the cabinet, draws its personnel and  
its power to govern from the legislative branch, the Parliament; and the cabinet  
controls the Parliament.27 However, in my view, this position is insufficiently  
nuanced because it ignores the realities of how responsible government functions  
in practice and the constitutional conventions that hedge that practice about.28  
26 New Brunswick Broadcasting Co. v. Nova Scotia (Speaker of the House of Assembly), [1993]  
1 S.C.R. 319, at p. 389, cited in Criminal Lawyers’ Association of Ontario, at para. 29.  
27 Hogg, at para. 9-12.  
28 These practical realities were well known to the Williams Commission, which took them into account.  
See below at paras. 132-37.  
Page: 41  
[121] Professor Dennis Baker disputes Hogg’s assertion that the separation of  
powers between the executive and legislative branches would “make little sense  
in a system of responsible government”.29 To the contrary, Baker states: “Far from  
being antithetical to responsible government… the executive-legislative separation  
is logically necessary for responsible government to work.”30 I agree with Baker.  
[122] Bitter historical experience, Baker notes, made “neither legislative nor  
monarchical absolutism… particularly appealing.” He explains:  
While the pre-Civil War experience with Charles I  
confirmed the fears of a king with absolute prerogatives,  
the subsequent experience with the Long Parliament  
raised serious doubts about legislative supremacy.  
Following Montesquieu, Blackstone understood this  
history as confirming the desirability of partial executive  
and legislative independence since “either total union or  
total disjunction would in the long run lead to tyranny.”31  
[123] This insight led to the development of the mixed polity of the Westminster  
model. Baker explains: “To fulfill its purpose of moderate government, the  
separation of powers might permit significant inter-branch interactions, even  
exertions of influence and control, but must prohibit arrangements that place one  
power entirely in the hands of another.”32  
29 Hogg, at para. 7-15, cited in Dennis René Baker, Not Quite Supreme: The Courts and Coordinate  
Constitutional Interpretation, (Montreal: McGill-Queens University Press, 2010), at p. 61.  
30 Baker, at p. 61.  
31 Baker, p. 58 (emphasis in original; footnotes omitted).  
32 Baker, p. 60. Apart from constitutional conventions, Baker notes, at pp. 61-62, that ss. 53 and 54 of the  
Constitution Act, 1867 apportion responsibilities over public finances between the executive and  
legislative branches, thus explicitly recognizing the separation of powers beyond the constitutional  
Page: 42  
[124] Baker argues that “the separation of powers continues to play a vital [role]  
in Canada‘s constitutional design”, albeit one that has been obscured by what he  
calls “the exaggerated claim of executive-legislative fusion”.33 “Viewed through this  
lens,” he notes, “the fundamental rule of the separation of powers (the power of no  
branch may be wholly exercised by another) can be easily discerned.”34 Each  
branch has a sphere of independence, but each is sufficiently hobbled to require  
the support of the others, which leads to a measure of interdependence. For  
example, the legislative branch has no executive capacity and the executive  
branch cannot enact legislation.35  
[125] The concept of fusion implies that executive control of the legislative branch  
is absolute, but this does not bear close scrutiny. It is more accurate to say that  
there is “a considerable degree of integration” between the legislative and  
executive branches.36  
[126] However, most telling, in my view, is Baker’s observation that: “[T]he subtle  
interplay of formal and informal power maintains and animates an effective  
conventions, citing Janet Ajzenstat, The Once and Future Canadian Democracy: An Essay in Political  
Thought (Montreal and Kingston: McGill-Queen’s University Press, 2003), at p. 65.  
33 Baker, at p. 83.  
34 Baker, at p. 83.  
35 See Baker, Chapter 3, especially pp. 61-63.  
36 Wells v. Newfoundland, [1999] 3 S.C.R. 199, citing Attorney General of Quebec v. Blaikie et al., [1981]  
1 S.C.R. 312, at p. 320. In Wells, the court also stated that even though the “separation of powers is not a  
rigid and absolute structure”, the court “should not be blind to the reality of Canadian governance that,  
except in certain rare cases, the executive frequently and de facto controls the legislature”: at para. 54.  
Page: 43  
institutional separation between the legislature and the executive.”37 This is the  
Anglo-Canadian version of constitutional checks and balances. It raises squarely  
the issue of Cabinet functionality; Cabinet serves as a “connecting link” between  
the two branches.38  
Cabinet, in functional terms  
[127] I now look more closely at the role of Cabinet within the Westminster system.  
In functional terms, Cabinet is to be understood as “a forum, presided over by the  
Prime Minister, where Ministers meet to propose, debate and decide government  
policy and action.39 It is “the place where Ministers decide, as a group, how the  
executive power should be exercised.40  
[128] Several building blocks are essential for Cabinet to be able to function  
effectively as a political body nested in Parliament or in the Legislative Assembly.  
These building blocks are fostered and protected by constitutional conventions.  
I focus on three: candour, solidarity, and confidentiality. Necessary and tight links  
among these conventions make possible the proper functioning of our  
37 Baker, at p. 83.  
38 The description of Cabinet as a “connecting link” is drawn from Walter Bagehot’s The English  
Constitution, 7th ed. (London: Kegan Paul, Trench, Trübner & Co., 1894), at p. 11. However, although  
I accept this connecting link concept, I reject Bagehot’s overall executive-legislative fusionist view.  
39 Yan Campagnolo, “The Political Legitimacy of Cabinet Secrecy” (2017) 51:1 R.J.T.U.M. 51, at p. 60.  
Campagnolo also explains, at pp. 60-61, that unlike the Privy Council, Cabinet has no legal existence or  
power. Rather, it is “an informal advisory body”. Executive power is “exercised by the Governor in Council  
or individual Ministers”, although “from a conventional perspective, the Governor in Council or individual  
Ministers act on the advice of the Cabinet.”  
40 Campagnolo, at p. 60.  
Page: 44  
parliamentary system in which the risk of a vote of no-confidence is ever-present.  
This risk is particularly acute in minority governments but still exists in a majority,  
if only as a more remote possibility. The Prime Minister and Cabinet must  
accommodate Cabinet’s own internal tensions, occasionally balky bureaucrats,  
hear from caucus and secure caucus support, marshal sufficient support in the  
House (challenging in minority times), and attune the government’s program both  
to day-to-day contingencies and to past and future electoral commitments  
designed to secure re-election.  
[129] Cabinet functionality depends on its members being free to communicate  
with complete candour. As McLachlin C.J. noted: “Those charged with the heavy  
responsibility of making government decisions must be free to discuss all aspects  
of the problems that come before them and to express all manner of views, without  
fear that what they read, say or act on will later be subject to public scrutiny”.41  
Cabinet could not carry out its policy-making and policy-vetting responsibilities if  
its members were inhibited in their debate by the prospect of public disclosure.  
[130] As for solidarity, all ministers accept responsibility collectively for Cabinet  
decisions and must resign or expect dismissal if they publicly dissent.42 Ministers  
could not credibly offer public support and positive explanations for policy  
41 Babcock v. Canada (Attorney General), 2002 SCC 57, [2002] 3 S.C.R. 3, at para. 18.  
42 Sir W. Ivor Jennings, Cabinet Government, 2nd ed. (Cambridge: Cambridge University Press, 1951),  
at pp. 257-58. See also Hogg, at para. 9-7.  
Page: 45  
decisions they opposed in Cabinet deliberations were that opposition to become  
publicly known.  
[131] Confidentiality links candour and solidarity. The confidentiality of Cabinet’s  
deliberations enables frank discussion and dissent during its meetings while  
preserving public-facing collective responsibility for its decisions.43 These three  
essential constitutional conventions underwrite the protected sphere in Cabinet  
within which government policy can be developed and debated, as the cases  
recognize.44  
(ii) The policy context for access to information  
[132] The Williams Report led to the enactment of the Act in 1988. The policy  
development process was sensitive to the political realities of Cabinet government  
and the functional issues because, to repeat, the terms of reference directed the  
Commission to consider changes that were “compatible with the parliamentary  
traditions of the Government of Ontario.45  
[133] The Williams Report recognized the tension between “a compelling public  
interest in open government”, on the one hand, and “a compelling public interest  
43 See Campagnolo, at p. 63, and a publication from the Privy Council Office that, among other things,  
summarizes the principles of Cabinet solidarity and confidentiality: Canada, Privy Council Office, Open  
and Accountable Government (2015), online: <http://pm.gc.ca/eng/news/2015/11/27/open-and-  
accountable-government>.  
44 In addition to the text above, see below at paras. 163-65.  
45 Williams Report, at p. 83.  
Page: 46  
in effective government”, on the other, which recognizes “the critical needs of  
government for confidentiality”.46 The Report did not set about to substantially  
depart from Cabinet’s current practices or well-established traditions, which are  
rooted in constitutional conventions, nor did its recommendations do so. The  
Report, as an added example, noted the need to preserve the anonymity of public  
servants,47 because not doing so “would mark a significant departure from this  
well-established tradition,48 which was not desired.  
[134] The goal was to provide public accessibility to government documents in  
order to facilitate “[i]ncreased access to information about the operations of  
government”.49 The Williams Report recognized that there was a “need to render  
government more accountable to the electorate”, and that “facilitating informed  
public participation in the formulation of public policy” was desirable.50 Achieving  
these ends would enhance the ability of the public to hold elected representatives  
accountable and the ability of members of the legislature to hold the executive  
accountable. The Report was confident that the “critical balance between the  
public interest in access and the government need for confidentiality” could be  
46 Williams Report, at p. 235.  
47 Williams Report, at p. 86.  
48 Williams Report, at p. 90 (emphasis added).  
49 Williams Report, at p. 77.  
50 Williams Report, at p. 77.  
Page: 47  
“achieved by means of statutory exemptions from the general rule of public  
access.51  
[135] Pertinent to the task of interpreting the exemption in s. 12(1) of the Act, the  
Williams Report stated: “[I]t is obvious that the confidentiality of Cabinet  
deliberations must be preserved in a freedom of information scheme”.52 The  
question was “how an exemption relating to this matter should be drafted”.53 The  
Report listed documents considered to be “Cabinet documents”:  
[A]gendas, informal or formal minutes of the meetings of  
Cabinet committees or full Cabinet, records of decision,  
draft legislation, Cabinet submissions and supporting  
material, memoranda to and from ministers relating to  
matters before Cabinet, memoranda prepared by  
Cabinet officials for the purpose of providing advice to  
Cabinet, and briefing materials prepared for ministers to  
enable them to participate effectively in Cabinet  
discussions.54  
[136] The Williams Report noted: “The disclosure of many of these documents  
would have the effect of disclosing the nature of Cabinet discussions and the  
advice given or received by Cabinet members”, and accordingly “all such material  
should be considered exempt under a freedom of information scheme.55  
51 Williams Report, at p. 277.  
52 Williams Report, at p. 285.  
53 Williams Report, at p. 285.  
54 Williams Report, at p. 285 (footnote omitted).  
55 Williams Report, at p. 285 (emphasis added).  
Page: 48  
[137] I now turn to the internal dimension of the context by which the Act as a  
whole must be given a coherent interpretation that reconciles its purpose of  
promoting access to information with the purpose of its s. 12(1) Cabinet records  
exemption.  
(b) The internal dimension  
[138] Two observations: First, it was open to the legislature to enact legislation  
requiring Cabinet to be much more forthcoming in its disclosure than Cabinet’s  
prior practices or well-established traditions would permit or require.56 Instead, the  
enacted legislation contains provisions that are substantially similar to those  
proposed in the relevant sections of the Williams Report. Second, while providing  
a right of access to government information, the Act shares the Report’s real  
diffidence around “Cabinet records”.  
[139] Against the rich background of the external dimension described above,  
I look at s. 12(1) from the perspective of the text and the legislative history, which  
together show the legislature’s intent.  
[140] The full text of s. 12(1) provides:  
(1) A head shall refuse to disclose a record where the disclosure  
would reveal the substance of deliberations of the Executive Council  
or its committees, including:  
56 This is what the Nova Scotia legislature did, as I discuss below at para. 162.  
Page: 49  
(a) an agenda, minute or other record of the deliberations  
or decisions of the Executive Council or its  
committees;  
(b) a  
record  
containing  
policy  
options  
or  
recommendations submitted, or prepared for  
submission, to the Executive Council or its  
committees;  
(c) a record that does not contain policy options or  
recommendations referred to in clause (b) and that  
does contain background explanations or analyses of  
problems submitted, or prepared for submission, to  
the Executive Council or its committees for their  
consideration in making decisions, before those  
decisions are made and implemented;  
(d) a record used for or reflecting consultation among  
ministers of the Crown on matters relating to the  
making of government decisions or the formulation of  
government policy;  
(e) a record prepared to brief a minister of the Crown in  
relation to matters that are before or are proposed to  
be brought before the Executive Council or its  
committees, or are the subject of consultations among  
ministers relating to government decisions or the  
formulation of government policy; and  
(f) draft legislation or regulations. [Emphasis added.]  
The meaning of the underlined words – “the substance of deliberations” and  
“including” – is hotly disputed in this case.  
(i) The text  
[141] The first perspective relates to the strength of the provision’s language. I  
infer that the exemption in s. 12(1) for Cabinet records is intended to be especially  
Page: 50  
strong. The text in the opening words of s. 12(1) is imperative: “A head shall refuse  
to disclose a record…” (emphasis added). This mandatory wording leaves the  
head no discretion. Similar language is found in other sections such as s. 21  
(personal privacy). Contrast this with the permissive and discretionary language  
about records covered in certain other sections, including s. 13(1) (advice to  
government), which the “head may refuse to disclose” (emphasis added). This  
inference about the strength of the s. 12(1) exemption is reinforced by contrasting  
s. 23 of the Act, which builds in flexibility and allows exemptions from disclosure to  
be lifted where there is a “compelling public interest”. It is especially instructive that  
the s. 23 public interest override does not apply to Cabinet records under s. 12(1),  
even though it does to a refusal under s. 13. Nor does the purpose language in  
s. 1(a) of the Act, which provides that “necessary exemptions from the right of  
access should be limited and specific”, take priority over the s. 12(1) exemption.  
(ii) The legislative history  
[142] The second perspective on the text takes into account the legislative history,  
which can provide guidance in statutory interpretation.57 The Attorney General for  
Ontario points out that a proposed amendment to Bill 34 (the predecessor draft Bill  
to the Act) would have limited the exemption from disclosure in s. 12(1) “solely” to  
57 See 1704604 Ontario Ltd. v. Pointes Protection Association, 2020 SCC 22, 449 D.L.R. (4th) 1, at  
paras. 12-14, citing Canada (Canadian Human Rights Commission) v. Canada (Attorney General),  
2011 SCC 53, [2011] 3 S.C.R. 471, at para. 43.  
Page: 51  
the records listed in the subparagraphs.58 The amendment was defeated. This  
reinforces the view that the expression the substance of deliberations” was meant  
to be broad enough to encompass the listed records without being limited to only  
those records.  
[143] I make four additional points. First, I set out above the Williams Report’s  
explanation for the list of particular records it would have included in the  
subparagraphs. However, here I focus on different words: “The disclosure of many  
of these documents would have the effect of disclosing the nature of Cabinet  
discussions and the advice given or received by Cabinet members.59 In other  
words, the Report’s focus was less on the list of records than on the principle:  
keeping confidential “the nature of Cabinet discussions” so that those discussions  
could proceed unharried by outside influences. In other words, even though not all  
the listed documents only many would or would always have the effect of  
disclosing Cabinet discussions, it is noteworthy that the Report recommended that  
all such material should be considered exempt under a freedom of information  
scheme.60  
58 Ontario, Legislative Assembly, Official Report of Debates (Hansard), 33rd Parl., 1st Sess., No. 113  
(10 February 1986), at p. 3955 (Norman Sterling).  
59 Williams Report, at p. 285 (emphasis added).  
60 Williams Report, at p. 285 (emphasis added).  
Page: 52  
[144] My second point is that it is instructive to contrast the wording of the Williams  
Report recommendation on Cabinet records with the opening words of s. 12(1)61:  
Freedom of Information and  
Protection of Privacy Act, R.S.O.  
1990, c. F.31, s. 12(1)  
Text of the Williams Report  
Cabinet records  
1. We recommend that the  
proposed freedom of information  
12(1) A head shall refuse to disclose a law contain an exemption for  
record where the disclosure would  
reveal the substance of deliberations  
of the Executive Council or its  
committees, including,  
documents whose disclosure would  
reveal the substance of Cabinet  
deliberations and, in particular, that  
the following kinds of Cabinet  
documents be the subject of this  
exemption:  
[145] Note the use of the words, “the following kinds of Cabinet documents” in the  
Williams Report. This is another way of expressing and underlining the Report’s  
worry about keeping “the nature of Cabinet discussions” confidential.  
[146] Third, the Williams Report gave two signal cautions that also made their way  
into the Act. Any disclosure regime must not have the effect of rushing Cabinet into  
a decision. Accordingly, there should be no disclosure of material forwarded before  
Cabinet’s consideration of it.62 Further, the Report accepted that “[t]here may be  
many situations in which Cabinet might properly wish to delay public  
announcements of its decisions.”63  
61 The full contrasting text is set out in the Appendix.  
62 Williams Report, at p. 287.  
63 Williams Report, at p. 286.  
Page: 53  
[147] My fourth point is that the Act expands the reach of the exemption beyond  
the Report’s recommendations in several ways. Notably, s. 12(1) of the Act  
generally uses the formulation “of the Executive Council or its committees” while  
the Report only used a similar formula once.64 Next, the Report would have limited  
the exemption under subparagraph (b) to “records containing proposals or  
recommendations submitted, or prepared for submission, by a Cabinet Minister to  
Cabinet”, but the Act provides a broader, more generic exemption: “policy options  
or recommendations submitted, or prepared for submission”. Finally, the Report  
would have limited the exemption under subparagraph (c) to the time “before such  
decisions are made” while the Act states: “before those decisions are made and  
implemented”. The enacted language is more protective of Cabinet records than  
the proposed language in the Williams Report.  
[148] I noted above that the Williams Report evinced real diffidence around the  
confidentiality of Cabinet records. The Commission appears to have favoured  
setting clear rules for Cabinet records.65 The practical reason for such a blanket  
64 Emphasis added throughout.  
65 The Report’s recommendations focused on the protection of physical Cabinet records. This concern  
with physical documents may relate to the historical formalization of Cabinet meetings. As Campagnolo  
notes, at pp. 72-77, prior to the 20th century, Cabinet meetings were informal affairs with no organized  
system of record-keeping. The only official document recording Cabinet discussions was a letter from the  
Prime Minister to the Sovereign. However, due to the increasing complexity of state activities, measures  
were taken to improve executive decision-making efficiency, including the introduction of Cabinet  
secretariats. Yet, the establishment of Cabinet secretariats was accompanied by a new risk: that the  
written records of Cabinet meetings could be potentially accessible, including by members of an incoming  
government following a change of power. This led to the development of conventions that focused on the  
protection of physical records of Cabinet’s deliberations.  
Page: 54  
rule is to avoid disputes over specific documents of the sort this case exemplifies.  
I will return to this point below.  
What is the purpose of s. 12(1) of the Act?  
[149] The consideration of the external and the internal contextual dimensions set  
out above leads me to conclude that the purpose for the exemption from the  
disclosure of Cabinet records in s. 12(1) of the Act is to establish a robust and well-  
protected sphere of confidentiality within which Cabinet can function effectively,  
one that is consistent with the established conventions and traditions of Cabinet  
government. “The preservation of the confidentiality of Cabinet discussions… [is]  
a necessary feature of a freedom of information scheme ‘compatible with the  
parliamentary traditions of the Government of Ontario’”, the Williams Report noted,  
warning that giving the public a right of access to documents revealing the nature  
of Cabinet deliberations would be a substantial departure from current practice.66  
[150] This purpose to establish a robust and well-protected sphere of  
confidentiality within which Cabinet can function effectively is reinforced by the  
mandatory and absolute nature of the protection in s. 12(1) and by the exclusion  
of s. 12(1) from the s. 23 public interest override. It is also more modestly reinforced  
by the slight adjustments in s. 12(1)’s subparagraphs in favour of more  
66 Williams Report, at p. 85.  
Page: 55  
confidentiality made by the legislature to the language proposed in the Williams  
Report.  
[151] At bottom, s. 12(1)’s purpose resonates profoundly with the values and  
virtues of Ontario’s version of Westminster responsible government, and facilitates  
what Baker called, to repeat: “the subtle interplay of formal and informal power  
[that] maintains and animates an effective institutional separation between the  
legislature and the executive.”67  
In light of the context and purposes, what interpretation of the  
text of s. 12(1) should the Commissioner have adopted?  
[152] The analysis of this question is divided into three sections: the approach to  
be applied to the interpretation of s. 12(1) of the Act; the role of the functional  
approach; and the Premier’s role in Cabinet in the interpretation exercise. In my  
view, the “illustrative approach” best captures the purpose of the exemption, and  
is supported by the functional approach to Cabinet government discussed above,  
taking into account the particular role played therein by the Premier.  
(a)Two approaches to the interpretation of s. 12(1) of the Act  
[153] When the word “including” is used in legislation, the issue often is which of  
two approaches, the “expansive approach” or the “illustrative approach”, was  
legislatively intended. Professor Ruth Sullivan states that: “The purpose of a list of  
67 Baker, at p. 83.  
Page: 56  
examples following the word ‘including’ is normally to emphasize the broad range  
of general language and to ensure that it is not inappropriately read down so as to  
exclude something that is meant to be included.68 Sullivan adds: “It is not always  
obvious whether a list that follows ‘includes’ is meant to expand the scope of the  
stipulated definition or merely illustrate it.”69  
[154] In interpreting s. 12(1), the Commissioner did not take the “illustrative  
approach”, which is endorsed by the Attorney General for Ontario. Instead he took  
the competing “expansive approach”, which is endorsed by the CBC and the  
interveners.  
[155] The expansive approach holds that, but for their express inclusion in s.  
12(1)’s subparagraphs, the listed records would not necessarily be caught by the  
opening words and so would otherwise require specific exemption from  
disclosure.70 The subparagraphs are thus said to “expand” the scope of the general  
exemption of records that “would reveal the substance of deliberationsby going  
beyond the underlined words to the list in the subparagraphs. The expansive  
approach takes a correlatively narrow view of the meaning of that expression.  
68 Ruth Sullivan, Sullivan on the Construction of Statutes, 6th ed. (Toronto.: LexisNexis Canada, 2014),  
at para. 4-38.  
69 Sullivan, at paras. 4-41 to 4-42, citing Entertainment Software Association v. Society of Composers,  
Authors and Music Publishers of Canada, 2012 SCC 34, [2012] S.C.R. 231.  
70 Reasons of Sossin J.A., at para. 43.  
Page: 57  
[156] Take the word “agenda” in subparagraph (a) as an example. Because an  
agenda sets out a list of topics for discussion at a meeting, it could not reveal any  
actual deliberations. The expansive approach argues that “agenda” should not be  
understood to be included in the expression “the substance of deliberations”  
because that expression refers only to actual deliberations at the Cabinet table.  
Or, as my colleague puts it, the Commissioner’s approach is only “protective of  
communications within Cabinet’s deliberative process that would reveal the  
substance of its formulation of government policies.”71 But for its inclusion in  
subparagraph (a), a Cabinet agenda would be disclosable under s. 1 of the Act.  
On the expansive approach, the same argument would apply to the records in  
subparagraphs (b), (c), (e), and (f), which are all records prepared by someone  
else for Cabinet to discuss and would not thereby describe any actual  
deliberations.  
[157] By contrast, the illustrative approach holds that subparagraphs (a) to (f)  
serve to identify or “illustrate” the types of records that, if disclosed, would reveal  
the substance of deliberations. The list of protected records in the subparagraphs  
informs the interpretation of “the substance of deliberations” and posits a different  
meaning: this expression refers to the nature of the topics, subject matters, or  
71 Reasons of Sossin J.A., at para. 37.  
Page: 58  
things Cabinet would be discussing, as well as to the deliberations themselves.  
The illustrative approach takes a broader view of the exemption.  
[158] In my opinion, the illustrative approach best achieves and instantiates the  
purpose of s. 12(1), which is to establish a robust and well-protected sphere of  
confidentiality within which Cabinet can function effectively. This result flows from  
the purpose of the legislation and the legislative history of s. 12(1) of the Act  
discussed above, and by the functional and purposive approach taken in the cases  
on the operation of the separated powers, and the particular role assigned to the  
Premier.  
[159] I am fortified in my view by the decision of the British Columbia Court of  
Appeal in Aquasource.72 That court took the same approach to Cabinet records as  
the Williams Commission, and adopted a broad reading of “the substance of  
deliberations” in s. 12(1) of B.C.’s legislation, which provided:  
The head of a public body must refuse to disclose to an applicant  
information that would reveal the substance of deliberations of the  
Executive Council or any of its committees, including any advice,  
recommendations, policy considerations or draft legislation or  
regulations submitted or prepared for submission to the Executive  
Council or any of its committees.73  
72 Aquasource Ltd. v. British Columbia (Freedom of Information and Protection of Privacy Commissioner),  
58 B.C.L.R. (3d) 61.  
73 Freedom of Information and Protection of Privacy Act, R.S.B.C. 1996, c. 16, s. 12(1).  
Page: 59  
[160] Donald J.A. held that the phrase “the substance of deliberations”, when read  
together with the clause, “including any advice, recommendations, policy  
considerations or draft legislation or regulations”, plainly refers to “the body of  
information which Cabinet considered (or would consider in the case of  
submissions not yet presented) in making a decision”.74 Accordingly, s. 12(1) “must  
be read as widely protecting the confidence of Cabinet communications.75 Donald  
J.A. concluded that the test for whether something is protected under s. 12(1) is:  
“Does the information sought to be disclosed form the basis for Cabinet  
deliberations?”76 In my view, the Aquasource approach is sound.  
[161] The Commissioner rejected this approach, preferring the analysis of the  
Nova Scotia Court of Appeal in O’Connor.77 I disagree with my colleague that the  
Commissioner made appropriate use of O’Connor. In that case, the court rejected  
Donald J.A.’s focus on the “body of information” Cabinet considered in its  
deliberations, and instead adopted a narrower test for Nova Scotia’s Cabinet  
records exemption.78 The Commissioner preferred the statutory interpretation of  
O’Connor to that of Aquasource on the basis that “the general approach articulated  
74 Aquasource, at para. 39 (emphasis added).  
75 Aquasource, at para. 41.  
76 Aquasource, at para. 48.  
77 O’Connor v. Nova Scotia (Deputy Minister of the Priorities & Planning Secretariat), 2001 NSCA 132,  
197 N.S.R. (2d) 154.  
78 O’Connor, at paras. 90-92.  
Page: 60  
by the Nova Scotia Court of Appeal in O’Connor… aligns more closely with the  
language of the [Ontario] exemption”.79  
[162] In applying the O’Connor approach to the words of s. 12(1) of the Ontario  
Act, the Commissioner failed to adequately take into account the salient  
differences between the respective statutes. In O’Connor, Saunders J.A. found  
that his province’s access to information legislation is uniquely and “deliberately  
more generous to its citizens and is intended to give the public greater access to  
information than might otherwise be contemplated in the other provinces and  
territories in Canada”, including Ontario.80 It is decidedly not the case that the  
O’Connor approach aligns more closely with the language of the [Ontario]  
exemption”. There is simply no basis on which the Commissioner could reasonably  
prefer O’Connor to Aquasource. Given the text, context, and purpose of Ontario’s  
Act, the opening words of s. 12(1) of the Act create a broader sphere of protection  
surrounding Cabinet confidentiality. This includes protection over “the body of  
information” Cabinet will consider in its deliberations.  
(b) The functional approach in interpretation  
[163] I now go deeper into the constitutional backdrop and draw on the cases  
describing what is necessary for the proper and effective functioning of Cabinet  
79 Order PO-3973, at para. 97.  
80 O’Connor, at para. 57.  
Page: 61  
government. The functional approach is evident in Ontario (Public Safety and  
Security) v. Criminal Lawyers' Association, where McLachlin C.J. and Abella J.  
make several pertinent observations.81 They note: “It may also be that a particular  
government function is incompatible with access to certain documents.” The  
example they give is the need to preserve secrecy and privacy in judicial  
deliberations. Public access “would impair the proper functioning of the court by  
preventing full and frank deliberation and discussion at the pre-judgment stage.”  
They add: “The principle of Cabinet confidence for internal government  
discussions offers another example.” They urge that attention be paid to the  
“historic function of a particular institution [which] may assist in determining the  
bounds of institutional confidentiality” because “certain government functions and  
activities require privacy”.82 They explain: “Certain types of documents may remain  
exempt from disclosure because disclosure would impact the proper functioning of  
affected institutions.”  
[164] La Forest J. stated in Carey v. Ontario: “I would agree that the business of  
government is sufficiently difficult that those charged with the responsibility for  
running the country should not be put in a position where they might be subject to  
81 Ontario (Public Safety and Security) v. Criminal Lawyers' Association, 2010 SCC 23, [2010]  
1 S.C.R. 815, at para. 40.  
82 Ontario (Public Safety and Security), at para. 40, citing Montréal (City) v. 29521366 Québec Inc.,  
2005 SCC 62, [2005] 3 S.C.R. 141.  
Page: 62  
harassment making Cabinet government unmanageable.”83 In making this  
statement, he cited Lord Reid’s trenchant observation in Conway v. Rimmer:  
[The premature disclosure of Cabinet confidences] would  
create or fan ill-informed or captious public or political  
criticism. The business of government is difficult enough  
as it is, and no government could contemplate with  
equanimity the inner workings of the government  
machine being exposed to the gaze of those ready to  
criticise without adequate knowledge of the background  
and perhaps with some axe to grind.84  
McLachlin C.J. added her agreement in Babcock and pointed out that: “[M]inisters  
undertake by oath as Privy Councillors to maintain the secrecy of Cabinet  
deliberations and the House of Commons and the courts respect the confidentiality  
of Cabinet decision-making.”85  
[165] Manageability and reasonable functionality underpin the functional  
approach taken by the courts. I note that Babcock is cited several times in the  
Commissioner’s reasons, mostly in reciting Cabinet Office’s submissions, but he  
ignores its teaching. Instead, he should have paid more respectful attention to the  
constitutional backdrop, as the Assistant Commissioner did in Order PO-1725.86  
(c) The Premier’s role in Cabinet  
83 Carey v. Ontario, [1986] 2 S.C.R. 637, 1986 7, at para. 50.  
84 Conway v. Rimmer, [1968] A.C. 910 (H.L.), at p. 952, cited in Carey, at para. 49.  
85 Babcock, at para. 18.  
86 Order PO-1725, [1999] O.I.P.C. No. 153.  
Page: 63  
[166] Neither the premiers nor the Prime Minister are expressly mentioned in the  
Constitution Act, 1867. Their constitutional roles and functions are conventional,  
not prescribed, which makes it essential to be careful in discerning what is at stake  
when a convention is touched. Neither the reason for, nor the proper reach of, a  
convention is necessarily completely obvious.  
[167] In Order PO-1725, Assistant Commissioner Tom Mitchinson provided an  
insightful articulation of the Premier’s role in Cabinet. The requester in that case  
sought access to the electronic and hardcopy appointment books of a named  
senior employee in the Premier’s office, whose “job title and employment  
responsibilities deal directly and primarily with policy formulation and the overall  
priority-setting and co-ordination of the government’s policy agenda.”87 Many  
entries were found to qualify for exemption from disclosure under the opening  
words of s. 12(1), although some were not.88  
[168] The Assistant Commissioner considered carefully the “constitutional  
conventions and traditions surrounding the role of the Premier in matters pertaining  
to Cabinet deliberations.89 The Assistant Commissioner reviewed the authorities  
on the conventions surrounding the Premier and Cabinet, and accepted the  
87 Order PO-1725, at para. 57.  
88 Order PO-1725, at paras. 61-64.  
89 Order PO-1725, at para. 50.  
Page: 64  
description of the integral role the Premier plays in the functioning of Cabinet  
presented by Dussault and Borgeat, who state:  
[Cabinet] is responsible for determining the ways and  
means of economic, social and cultural progress and is  
called upon to translate into legislation and into concrete  
programs the values underlying its rise to power or its  
remaining in power. Above all, therefore, it represents a  
centre for reflection and decision. By its very nature, the  
Cabinet is an institution for compromise, with respect to  
which its primary role is to determine priorities, to plan  
and to establish political strategy.  
[T]he ultimate responsibility for decision-making,  
although ascribable to Cabinet members as a group, is  
conferred in particular upon the Prime Minister who  
dominates its activities. This results since he or she is the  
head of Cabinet and receives technical briefs and also  
since he or she has the power to determine the agenda  
for meetings and to exert control over the support staff.  
The Prime Minister has recently been termed “the guiding  
force, co-ordinator and arbitrator of the executive  
decision-making process”. Possessing, inter alia, such  
powers as the authority to appoint his or her colleagues,  
the Prime Minister dominates the administrative  
machinery.90  
Dussault and Borgeat noted that, while ministers are generally viewed as equals,  
the Prime Minister or Premier is “without doubt ‘a little more equal’ than the  
others”.91  
90 René Dussault and Louis Borgeat, Administrative Law, A Treatise, 2nd ed. (Toronto: Carswell, 1985),  
at pp. 59-60 (footnotes omitted), cited in Order PO-1725, at para. 52.  
91 Dussault and Borgeat, at p. 61. For additional discussion of Cabinet, see generally Dussault and  
Borgeat, at pp. 51-63.  
Page: 65  
[169] Against this background, the Assistant Commissioner framed three broad  
principles that guided his disposition, which warrant reproduction:  
Firstly, by virtue of the Premier’s unique role in setting the  
priorities and supervising the policy making, legislative  
and administrative agendas of Cabinet, the deliberations  
of the Premier, unlike those of individual ministers of the  
Crown, cannot be separated from the deliberations of  
Cabinet as a whole. The Premier’s consultations with a  
view to establishing Cabinet priorities are an integral part  
of Cabinets substantive deliberative processes. To the  
extent that records reflect consultations bearing on the  
policy making and priority setting functions within the  
constitutionally recognized sphere of the Premier’s  
authority as first minister, those records, by definition,  
may be seen as reflecting the substance of deliberations  
of the whole Cabinet.  
Secondly, in our modern parliamentary democracy, the  
Premier functions by and large through the  
instrumentality of staff within his Office.  
Thirdly, the Premier’s policy-making and priority setting  
functions do not occur in a vacuum, but within the political  
framework which brought the ruling party to power.  
Cabinet, and the Premier in his capacity as leader of the  
winning party, are charged with the task of prioritizing and  
implementing the major policy choices of party members  
by translating political party values into strategies for  
legislation and other programs. By virtue of his dual role  
as party leader and head of Cabinet, the Premier is at the  
apex of both the political and legislative policy-making  
functions. In the person of the Premier, Cabinet  
deliberations cannot be divorced from the consensus  
Page: 66  
building process that must occur within the democratic  
political environment.92  
[170] These words reveal the radical discontinuity between the approach taken in  
Order PO-1725 and by the Commissioner in this case. The Assistant  
Commissioner unequivocally found that, owing to the constitutional conventions  
and traditions, “the deliberations of the Premier, unlike those of individual ministers  
of the Crown, cannot be separated from the deliberations of Cabinet as a whole.”  
The Commissioner quotes this statement early in his reasons in reciting the  
submissions of Cabinet Office,93 but he never directly engages with the statement’s  
implications for the interpretation of s. 12(1).  
[171] The Commissioner’s chain of reasoning rests on two propositions. First,  
s. 12(1) implicitly distinguishes between the “substance of deliberations” and the  
outcome of deliberations. Second, s. 12(1) of the Act applies only to “Cabinet as  
a whole”; because the mandate letters are at best outcomes of the Premier’s  
deliberations, they do not fall within the expression “substance of deliberations”. In  
addition, the Commissioner construed Order PO-1725 too narrowly. I address  
each point in turn.  
(i) The distinction between outcomes and deliberations is not  
material in this case  
92 Order PO-1725, at paras. 54-56 (emphasis added).  
93 Order PO-3973, at para. 23.  
Page: 67  
[172] The Commissioner’s view is that the mandate letters are at best outcomes”  
of the Premier’s deliberations that do not fall within the “substance of deliberations”  
of Cabinet as a whole”.94 I reject the relevance of this distinction in this case.  
[173] I accept that there are circumstances where the distinction between the  
“substance” and the “outcome” of deliberations would be meaningful, such as  
when the outcome of Cabinet’s deliberations on an issue has been publicized but  
“the substantive details of the matters deliberated upon by Cabinet” to reach that  
outcome have not.95 However, I would qualify the distinction in two ways.  
[174] First, the Act is concerned about premature disclosure. For example,  
subparagraph (c) expressly exempts materials used in reaching a decision “before  
those decisions are made and implemented”. Further, s. 18(1)(g), exempts:  
“information… where the disclosure could reasonably be expected to result in  
94 I also note that the Commissioner followed a line of IPC cases that draws a distinction between the  
“substance” and the “subject matter” of deliberations: see Information and Privacy Commissioner of  
Ontario, Order PO-3719 (2017), at para. 42; Order PO-3720; Ontario (Ministry of Finance) (Re), [2017]  
O.I.P.C. No. 58, at paras. 33, 42, and 44; Interim Order MO-2964-I; Greater Sudbury (City) (Re), [2013]  
O.I.P.C. No. 254, at paras. 37-39, 43, and Interim Order MO-3684-I; North Bay (City) (Re), [2018] O.I.P.C.  
No. 236, at paras. 18-21. As I indicated above, I reject this distinction because, understood in its proper  
context, s. 12(1) aims to protect the confidentiality of certain kinds of documents whose disclosure would  
reveal the matters Cabinet would be discussing, not just the content of its discussions. I also note that  
much of this IPC case law is rooted in the interpretation of the meaning of “the substance of deliberations”  
as it appears in s. 6(1)(b) of the Municipal Freedom of Information and Protection of Privacy Act. R.S.O.  
1990, c. M.56. This section permits a head to refuse to disclose a record “that reveals the substance of  
deliberations of a meeting of a council, board, commission or other body or a committee of one of them if  
a statute authorizes holding that meeting in the absence of the public” (emphasis added). Although both  
provisions use the phrase “the substance of deliberations”, the s. 6(1)(b) exemption is unrelated to  
Cabinet records and therefore does not engage the constitutional conventions surrounding Cabinet  
confidentiality. As a result, the case law interpreting the scope of s. 6(1)(b) is of limited use in discerning  
the scope of s. 12(1).  
95 Order PO-3752; Ontario (Ministry of Energy), [2017] O.I.P.C. No. 145, at para. 40.  
Page: 68  
premature disclosure of a pending policy decision”. In construing the Act  
coherently, the concern about premature disclosure counsels caution in making an  
overly sharp distinction between deliberations and outcomes in a chain of  
reasoning in assessing the application of s. 12(1).  
[175] Second, chains of deliberative reasoning are usually comprised of  
alternating outcomes and further deliberations until the end of the chain when the  
ultimate outcome appears. There is no sense in which the Act would require the  
disclosure of any “interim” outcome in a chain of deliberative reasoning. In Cabinet,  
where that deliberative chain culminates in draft legislation or regulations, it is still  
protected from disclosure.  
[176] The question arises: If the mandate letters are disclosable on the basis that  
they are merely the outcomes of the Premier’s deliberations and are not therefore  
part of Cabinet deliberations, what other decisions of the Premier sent in  
documentary form to Cabinet ministers would not also be disclosable? The  
ramifications could force large and ultimately unproductive changes in the way the  
Premier communicates with ministers. Consider, for example, the content of a new  
mandate letter written to a minister just appointed to replace an underperforming  
minister. It is hard to imagine that such a new letter would not reflect in its  
instructions to the new minister the Premier’s displeasure with the performance of  
Page: 69  
the old minister, quite likely the subject of Cabinet discussion, especially when  
compared to the mandate letter to the old minister.  
(ii) The Premier is not separate from Cabinet  
[177] As the above discussion of Order PO-1725 reveals, it is a novel proposition  
one that I reject in this case that the Premier’s deliberations as head of Cabinet  
can be separated from those of the rest of Cabinet, specified by the Commissioner  
several times as “Cabinet as a whole”, for the purpose of applying the Act.96  
[178] Drawing a hard line between the Premier’s deliberative process and that of  
the rest of Cabinet would not respect the way Cabinet functions because it would  
interfere with “the subtle interplay of formal and informal power [that] maintains and  
animates an effective institutional separation between the legislature and the  
executive.”97 Doing so would be contrary to the instructions given to the Williams  
Commission, and faithfully reflected both in its Report and in the Act, that reforms  
be “compatible with the parliamentary traditions of the Government of Ontario.98  
96 The Premier is not generally set apart from Cabinet. As Hogg notes: “Not only do conventions  
presuppose the existence of law, much law presupposes the existence of conventions.The Constitution  
Act, 1867 was drafted the way it was because the framers knew that the extensive powers vested in the  
Queen and Governor General would be exercised in accordance with the conventions of responsible  
government, that is to say, under the advice (meaning direction) of the cabinet or in some cases the  
Prime Minister. Modern statutes continue this strange practice of ignoring the Prime Minister (or provincial  
Premier) and his cabinet. They always grant powers to the Governor General in Council (or the  
Lieutenant Governor in Council) when they intend to grant powers to the cabinet. The numerous statutes  
that do this are of course enacted in the certain knowledge that the conventions of responsible  
government will shift the effective power into the hands of the elected ministry where it belongs”:  
at para. 1-14 (footnote omitted).  
97 Baker, at p. 83.  
98 Williams Report, at p. 83.  
Page: 70  
[179] The Premier sets the “agenda” of the government and Cabinet in the large  
sense.99 His mandate letters reflect the outcome of a deliberative process on the  
Premier’s part, assisted no doubt by staff and political advisers. However, as  
Cabinet Office put it before the Commissioner, the letters also “initiate a continuing  
deliberative process at Cabinet”.100 They signal the tasks the agenda that the  
Premier expects each minister to undertake within the minister’s portfolio. In my  
view, the Premier’s deliberations leading to the mandate letters, and the letters  
themselves, are part of Cabinet’s deliberative process. The entire set of mandate  
letters should be seen as the starting instructions for Cabinet in the new mandate,  
or as “blueprint[s] to inform discussion at the Cabinet table.101 While they contain  
some campaign-style language, to varying degrees they also go further and, in  
some instances, signal the need for further policy work that will inevitably return to  
Cabinet. They are records that would reveal the nature of Cabinet deliberations  
within the meaning of “the substance of deliberations”.  
(iii) The Commissioner overstates the holding in Order PO-1725  
[180] The Commissioner overstates the holding in Order PO-1725 in asserting  
that the records at issue in that case “were deliberative in nature because they  
99 I do not use the word “agenda” in the technical meaning given by the Act: see Order PO-1725,  
at para. 60.  
100 Order PO-3973, at para. 27.  
101 Order PO-3973, at para. 27.  
Page: 71  
provided a roadmap revealing how and why policy choices were made by the  
Premier.”102  
[181] The holding in Order PO-1725 does not go that far. First, the Assistant  
Commissioner noted: “While many of these references consist of abbreviations,  
acronyms or initials, persons knowledgeable in the affairs of government would  
likely be in a position to identify most of these references both as to subject matter  
and the persons or entities involved.”103 He added:  
To the extent that the records reveal the issues and  
options upon which the Premier or the named individual  
is reflecting in formulating and establishing Cabinet’s  
“agenda” – used here in its broadest sense these  
records would tend to reveal the substance of this  
deliberative process and, therefore, the substance of the  
deliberations of Cabinet in the context of the Premier’s  
unique role within that body.104  
[182] Note that the “substance of the deliberative process” can only mean the  
subject matter under consideration, not the Premier’s actual deliberations. The  
Assistant Commissioner noted: “It is only by virtue of the capacity of these entries  
to reflect the Premier’s deliberations in establishing Cabinet’s priorities that they  
fall within the introductory wording of section 12(1) by revealing the substance of  
that exercise.”105  
102 Order PO-3973, at para. 130.  
103 Order PO-1725, at para. 58.  
104 Order PO-1725, at para. 59.  
105 Order PO-1725, at para. 60.  
Page: 72  
[183] There is, with respect, no way that the scheduling entries could reveal the  
Premier’s actual deliberations, only their subject matter. The entries would provide  
a form of roadmap as to the activities of the named employee, but they would not  
reveal “how and why policy choices were made by the Premier.” The  
Commissioner’s conclusion is completely speculative.  
[184] To conclude, because the mandate letters are the product of the Premier’s  
deliberations, in his role as head of Cabinet, and initiate Cabinet’s mandate going  
forward, they are, perforce, part of Cabinet’s deliberations and are fully protected  
from disclosure by the opening words of s. 12(1). As I stated at the outset, this  
determination is dispositive of the appeal.  
(d) The test applied  
[185] The design of the s. 12(1) exemption aims to protect the confidentiality of  
certain kinds of documents whose disclosure would reveal the nature of Cabinet’s  
deliberations, that is, the topics, subject matters, things or the body of information  
Cabinet would be discussing.  
[186] The pertinent question is whether the particular record resembles or is  
analogous to a record in the list or would otherwise reveal the nature of Cabinet  
deliberations. The mandate letters are analogous to the records listed in  
ss. 12(1)(d) and (e), respectively: “a record used for or reflecting consultation  
among ministers of the Crown on matters relating to the making of government  
Page: 73  
decisions or the formulation of government policy”, there being no doubt that the  
Premier is a Cabinet minister; and “a record prepared to brief a minister of the  
Crown in relation to matters that are before or are proposed to be brought before  
the Executive Council or its committees”. It is no stretch to apply these words to  
the mandate letters.  
[187] The design of the s. 12(1) exemption emplaces metaphorical bollards in the  
form of categorical exemptions in order to provide robust protection of the sphere  
of confidentiality within which Cabinet can function effectively. Departing from this  
interpretation would engage the Commissioner and the court in a deconstructive  
exercise in which every questioned record would be parsed and pared down to  
some irreducible core of actual communications at the Cabinet table. The  
Commissioner’s analysis portends this outcome.  
Is the Commissioner’s new test compatible with s. 12(1)?  
[188] At the outset, I stated that I would also allow the appeal on the basis that the  
Commissioner’s new test for an exemption from disclosure is fundamentally  
incompatible with the text, context, and purpose of s. 12(1) of the Act. I address  
the general approach to establishing an exemption, the Commissioner’s new test,  
and problems with the new test and why it is unreasonable.  
(a) Establishing an exemption from disclosure  
Page: 74  
[189] It is axiomatic that the party invoking the s. 12(1) exemption must explain  
why it applies. To some extent, that explanation will lay out the nature of the record  
and how it relates to s. 12(1), both the opening words and any applicable  
subparagraph. The forensic pattern is seen in the old case of Conway, which was  
a civil action by a former probationary police constable against his former