The Nature and Limitations of Administrative Review

The Nature and Limitations of Administrative Review

Introduction

An Administrative Review is a decision-making mechanism that is most often relied upon by centralized Canadian Forces (CF) statutory decision-makers at National Defence Headquarters (NDHQ).  Any member of the CF who has faced compulsory release from the CF due to conduct or performance deficiencies, medical employment limitations (MEL), or misconduct will likely be familiar with Administrative Review.  They will also likely no longer be a member of the CF.

The purpose of this Blog article is to introduce the reader to the framework and function of Administrative Review under Defence Administrative Order and Directive (DAOD) 5019-2, and describe its role in the broader administration of the affairs of the CF.  In particular, we will begin to look at the relationship between Administrative Review and other adjudicative processes created by, and under the authority of, the National Defence Act (NDA).  We will examine where Administrative Review is an effective decision-making mechanism, and where it has limitations or deficiencies.

This discussion will then set conditions to examine specific shortcomings of Administrative Review when it is used as a substitute for Code of Service Discipline proceedings or Harassment investigations, and when they are conducted following the assignment, by Director of Medical Policy (D Med Pol), of Medical Employment Limitations (MEL).

 

Description of Administrative Review

The principal policy instrument for Administrative Review is DAOD 5019-2.  What is crucial to understand is that it is not the sole adjudicative mechanism available to CF statutory decision-makers, nor is it even the most versatile.  In fact, in terms of fact-finding or weighing competing evidence, it is one of the least versatile.  The decision-maker is far removed from the subject or respondent of the Administrative Review, even though this ‘distance’ does not engender any real independence in decision-making.  The adjudicative mechanism is exclusively a ‘paper hearing’ (notwithstanding that, in the ‘digital age’, the file is just as likely to be a PDF file as a paper file) – it consists largely of what would otherwise be characterized as a ‘file review’.  There are few, if any, investigative protocols arising in DAOD 5019-2 and little scope for evaluating reliability and credibility when conflicting versions of events are presented.

The Table at Article 4.5 of DAOD 5019-2 identifies the types of Administrative Reviews that may be conducted, as well as the appropriate approving authority (AA) and the staff that will provide the analyst, based upon the relevant component of the CF and the rank of the respondent.

These categories include:

1. Performance deficiencies;
2. Prohibited drug use or other involvement with drugs, sexual misconduct, alcohol misconduct or any other conduct deficiency;
3. Medical Employment Limitations (MEL);
4. Terms of Service;
5. Non-effective Strength (limited solely to the Reserve Force); and
6. Performance deficiency regarding an untrained CF member.

These are broad categories, which offers a degree of flexibility, while the table provides precision regarding the appropriate decision-making authority.  Unfortunately, there is a tendency by CF policy-makers to amend these requirements, from time to time, by issuing direction an a Canadian Forces General Message (CANFORGEN) rather than amending the actual DAOD.  Thus, someone seeking policy direction from the DAOD (the appropriate, notoriously published policy instrument) may be unaware of an ‘amendment’ in the CANFORGEN.  Depending upon their circumstances, the CANFORGENs may not be readily accessible to them.

For example, CANFORGEN 049/19 establishes a new definition of ‘sexual misconduct’ and clarifies remedial processes, including amending DAOD 5019-5 Sexual Misconduct and Sexual Disorder.  There is no hyperlink in this blog to CANFORGEN 049/19 because it, and CANFORGENs generally, are not publicly available.  They are available on the Defence Wide Area Network (DWAN), which constitutes the CF’s unclassified intranet.  Nor is there any mention in DAOD 5019-5 that it is amended by a CANFORGEN.

In an earlier blog (on a separate website), I describe some of the pitfalls in using CANFORGEN – which is a messaging mechanism, and not a policy instrument – to create or amend policy.  CANFORGEN are useful tools for announcing policy.  After all, the intended purpose of CANFORGENs is to facilitate CF-wide announcements.  But CANFORGENs were not intended as policy instruments, and, for the reasons announced in the aforementioned Blog article, it is idiotic to use them as such.

Perhaps the most common AA for Administrative Reviews is Director Military Careers Administration (DMCA).  Certainly, most Administrative Reviews in which there is a likelihood or possibility of compulsory release being ordered under Items 2, 3, or 5 of the Table to QR&O article 15.01, will be conducted by DMCA and the various staff sections under DMCA’s supervision.  I use the terms ‘likelihood’ and ‘possibility’ optimistically.  Ideally, the determinations in Administrative Reviews would be the result of open-minded consideration of all relevant policy and evidence, resulting in a procedurally fair and substantively reasonable decision.  Regrettably, in my experience, this is often not the case, and there are often elements of closed-minded and ‘pre-determined’ decision-making.

An Administrative Review is best suited to making policy-led decisions based upon prior, reliable, adjudications conducted by disciplinary or administrative tribunals.  The ‘AA’ will apply established facts to an applicable policy regime and make a determination in the public interest, respecting the rights, interests, and privileges of the CF member who is subject to the decision (i.e. the respondent).  In these circumstances, centralized decision-making permits consistent and, ideally, rational decisions, while maintaining an open-minded approach.

An Administrative Review is often the ‘final’ administrative decision-making process (short of the adjudication of a grievance) in the administration of the affairs of the Canadian Forces.  Typically, Administrative Reviews will be conducted prior to making a decision to terminate a CF member’s service or to take significant remedial or administrative action against a CF member (such as compulsory occupational transfer or Counselling and Probation under DAOD 5019-4).  As we know from the first Blog article in this series, the 5019 series of DAOD address a variety of topics relating to conduct and performance deficiencies, which may (and in some cases must) be subject to adjudicative mechanisms before CF decision-makers rely on DAOD 5019-2.

Consequently, an Administrative Review is most effective when it relies upon prior adjudication under processes including: the Code of Service Discipline; DAOD 5012-0 Harassment Prevention and Resolution; Remedial Measures under DAOD 5019-4; or other administrative tribunals designed to gather evidence, provide procedural fairness, and support adjudicative decision-making.  These processes – which I would characterize as ‘originating processes’ – ensure that the respondent to allegations of misconduct or sub-standard performance is provided with notice of the allegations against them, adequate and timely disclosure of any supporting evidence, in order that they may offer full answer and defence or response to the allegations, to (hopefully) an unbiased and open-minded decision-maker.

The Administrative Review process described at DAOD 5019-2 does have specific measures, principally under Section 5 of the DAOD, that are clearly intended to provide a modicum of procedural fairness – or, at the very least, a semblance of procedural fairness.  However, the inclusion of steps supporting notice, disclosure, an opportunity to make representations, and the provision of reasons in support of a decision are not guarantees of a fair process.  A capacity to conduct sufficient enquiry, to weigh evidence, and to facilitate meaningful representations is also crucial.

Administrative Reviews, conducted by centralized CF statutory decision-makers, such as DMCA, can be effective in ensuring that decisions are made based upon a consistent application of CF policies.  The centralized nature of the decision-making paradigm serves as a quality control measure.  However, Administrative Reviews have proven to be remarkably incapable of making determinations when facts are in dispute or where there is an absence of prior adjudicative framework.  This is particularly problematic where those prior adjudicative mechanisms are mandatory for specific actions and CF personnel have a legitimate expectation that they will be used by CF decision-makers.  Administrative Reviews are most effective when relying upon prior adjudications by decision-makers who are better placed or better designed to make determinations or reliability and credibility or where investigative steps are required.

An Administrative Review is neither a sufficiently robust investigative process, nor is it an adequate adjudicative process where competing versions of events or evidence must be weighed and determined.  An Administrative Review can be used where an ‘AA’ is required to decide upon the appropriate recourse arising from a previously established performance or conduct deficiency, MEL, or even the end of Terms of Service.  However, the Administrative Review is an ineffective tool for making many underlying adjudications.  It is not designed to do so.  For example, where a CF member is subject to MEL imposed by D Med Pol, it is not DMCA who determines what the MEL should be.  DMCA determines what should be done in light of the MEL assigned by D Med Pol.

 

Administrative Reviews are not Investigative Mechanisms

An Administrative Review is not an investigative mechanism.  Notwithstanding that DAOD 5019-2 does establish a decision-making process, there is no guidance in the DAOD regarding investigative techniques or methods.  Arguably the sum total of the investigative direction for Administrative Review is provided in ‘Steps 1 to 4’ of the Table to article 5.3 of the DAOD:

1. The authority who recommends an AR must forward the recommendation and all supporting evidence to the appropriate AR analyst.
2. The AR analyst creates an AR case file.
3. The AR analyst sends an advisory message to the CAF member through the CAF member’s CO.
4. The AR analyst obtains the significant incident report, military police investigation report, board of inquiry report or other relevant information, as applicable.
5. If retention subject to employment limitations under the minimum operational standards related to universality of service is being considered, the AR analyst obtains military occupation information, for example, Annual Military Occupation Review information, rank documentation, etc., from the applicable authority.

That’s it.  There is no direction about any further inquiry.  There is nothing done to ‘fact check’ any information provided by the respondent’s chain of command.  ‘Step 6’ directs the analyst to: review all documentation; prepare a case file synopsis; and prepare a recommendation to be reviewed and approved by the chain of command.  There is no direction to ensure completeness of the information or to follow up where there may be gaps.  It is essentially a checklist.  And, as we will discuss below, that can lead to a lack of all relevant information.

Certainly, an Administrative Review is demonstrably not an investigative tool in the same manner as a disciplinary investigation under Chapter 106 of the QR&O or an Harassment Investigation under DAOD 5012-0 and the CF Harassment Prevention and Resolution Instructions.

Ultimately, the analyst relies upon the information conveyed by the ‘recommending authority’ (described under article 5.2 of the DAOD).  The analyst eventually obtains any relevant information and documents maintained by the Canadian Forces.  But what is most noteworthy about the information compiled for an Administrative Review is the reliance on reports such as military police reports or (although not expressly identified in the DAOD) unit disciplinary investigations (UDI).  These are investigations compiled principally for disciplinary purposes.  While there is no absolute prohibition on their use for administrative purposes, reliance upon untested disciplinary investigations for Administrative Reviews can be problematic.

In a recent matter in which I have been involved, an unnamed DMCA analyst justified a second Administrative Review (where a prior Administrative Review was concluded without any adverse action against the respondent) because of ‘new’ evidence that allegedly came to light.  But consider what the analyst said about the ‘discovery’ of this new evidence:

It was learned that while [the prior Administrative Review] was ongoing, the unit CO requested that a Unit Disciplinary Investigation (UDI) be conducted … . When the [earlier] AR decision was rendered … DMCA 2 staff was unaware that a UDI had been concluded [two years previously].  Subsequently, the unit CO sent the UDI report with the following information: …

Essentially, the DMCA analyst acknowledged that DMCA staff do not conduct independent investigation, or even fact check or query the information that is provided to them under ‘Step 1’ of the Administrative Review Process.  It simply relies on the information sent from the initiating unit, without querying its completeness, and, perhaps, on a rudimentary collation of basic career information.

As an aside, the statement by the unnamed DMCA analyst was ambiguous.  Based upon how the information was conveyed, it would appear that the CO ordered the conduct of a UDI “while” the prior Administrative Review was being conducted.  That is not actually what transpired in that case.  The UDI was conducted, over the course of 2 to 3 weeks, 2 years and 3 months prior to the conduct of the first Administrative Review.  However, for reasons that remain unknown, the contents of the UDI were not disclosed to DMCA, notwithstanding that the UDI was conducted by the same unit that had requested the Administrative Review.  Thus, even though the CO who ordered the Administrative Review had the UDI in his possession, it was not initially disclosed to DMCA.

DMCA analysts predicate their recommendations, and DMCA predicates her decisions, on a file that is prepared based upon what can best be characterized as a ‘checklist approach’ to decision-making.

 

Boiler-Plate Administrative Reviews are Ill-Suited to Weighing Contested Evidence

Administrative Reviews have limited capacity to make a qualitative assessment of ‘evidence’ and certainly no capacity to test credibility where there is conflicting evidence.  It is one thing for DMCA to rely upon the outcomes of adjudicative processes like courts martial, summary trials, or even harassment investigations and adjudications by a Responsible Officer (RO).  In all of those cases, determinations are made following robust investigation and consideration of evidence in which an accused or respondent is able to offer ‘full answer and defence’ to allegations.  It is quite different for DMCA to rely upon untested UDI or Military Police reports, which are generated for disciplinary processes that are never followed, and for which the ‘accused’, who is either never charged or never prosecuted, is denied an opportunity to offer full answer and defence, including an opportunity to test the reliability and credibility of the evidence against him.

Instead, DMCA analysts assert boiler-plate mantras like:

The standard of proof in administrative cases is different from that applied in criminal cases (beyond a reasonable doubt). The accepted standard of proof in administrative cases is the civil standard of proof based on the balance of probability (is it more likely than not that the alleged incident occurred, and what are the facts and justifications supporting it?). A criminal conviction or even a charge is not a prerequisite for a determination of misconduct. The information available must support such a determination based on a balance of probability. It must be noted that an administrative measure is not a punishment and may be ordered in conjunction with disciplinary action.

This slavish repetition of the civil burden of proof is treated as an absolute justification for drawing whatever conclusions DMCA wishes to draw.  But while the assertion of the civil burden of proof is repeated no fewer than three times in the boiler-plate language of Administrative Review disclosure, what is omitted in illuminating.

Under DAOD 5019-2, article 5.6 describes the requisite standard of proof for specific types of Administrative Review.  It states that:

… in an AR which may result in:

• • release under Item 1(b), 1(d), 2, 5(d) or 5(f) in the Table to QR&O article 15.01, Release of Officers and Non-Commissioned Members;
  • reversion in rank; or
  • a finding that the CAF member engaged in any activity described in an offence:
    • in the Code of Service Discipline; or
    • in any other federal legislation.

The evidentiary standard is “clear and convincing evidence”.

 

The enumerated release items are: ‘Service Misconduct’; ‘Fraudulent Statement on Enrolment’; ‘Unsatisfactory Service’; Not Advantageously Employable’; and, ‘Unsuitable for Further Service’.  All of these release items are predicated upon blameworthy conduct.  The required evidentiary standard is reflective of the unanimous judgment of the Supreme Court of Canada in F.H. v McDougall, [2008] 3 SCR 41, per Rothstein J at para 46:

… evidence must always be sufficiently clear, convincing and cogent to satisfy the balance of probabilities test.  But again, there is no objective standard to measure sufficiency.  In serious cases, like the present, judges may be faced with evidence of events that are alleged to have occurred many years before, where there is little other evidence than that of the plaintiff and defendant.  As difficult as the task may be, the judge must make a decision.  If a responsible judge finds for the plaintiff, it must be accepted that the evidence was sufficiently clear, convincing and cogent to that judge that the plaintiff satisfied the balance of probabilities test.

Justice Rothstein was highlighting that, in administrative proceedings (barring a statutorily imposed threshold such as might be found in ‘disciplinary’ proceedings under certain Police Services Acts[1]) there is one burden of proof.  He rejected what he referred to as the “shifting standard” of proof based upon the seriousness of the matter.  What Justice Rothstein described was the examination of the quality of the evidence available to the decision-maker, particularly when making a determination of blameworthy conduct.

This vital consideration is mentioned in DAOD 5019-2.  It is absent from the boiler-plate Administrative Review analysis by DMCA staff.  [And, as an aside to DMCA staff: simply inserting rote boiler-plate language reflective of the judgment in F.H. v McDougall is not sufficient to ‘review-proof’ DMCA decisions.  There still has to be some sort of cogent analysis and sufficiency of reasons].

DAOD 5019-2 also provides a description, at article 5.12 and 5.13, of the purposes served by the reasons that must be given in a decision based upon an Administrative Review, and the scope of what the reasons must include:

5.12 The reasons for imposing an administrative action in respect of a CAF member must contain sufficient information:

1. 1. to enable the CAF member to understand why the administrative action was imposed; and
   2. to permit review by grievance authorities.

5.13 The reasons provided must specifically:

1. 1. identify the applicable evidence;
   2. explain how the evidence was treated; and
   3. state any findings based upon the evidence, and given those findings, explain why the specific administrative action was imposed.

This description reflects the purpose and scope of reasons described in R v Sheppard, 2002 SCC 26, [2002] 1 SCR 869 (concerning reasons in a criminal prosecution) and Baker v Canada (Minister of Citizenship and Immigration), [1999] 2 SCR 817 (regarding administrative adjudication, where there is a significant impact on the rights, interests, and privileges of the affected person).  Unfortunately, although DAOD 5019-2 makes rudimentary reference to these thresholds, it does not provide much direction to decision-makers regarding how to ensure these thresholds are met.  There is no direction regarding how evidence should be weighed or how the decision-maker can explain why certain evidence is accepted and other evidence is rejected.  More unfortunate still, based upon my experience with Administrative Review disclosure and decisions while I have been in private practice, there is little indication that DMCA and her analysts follow even this rudimentary direction.

I have previously described the deficiencies of Administrative Review processes, particularly in the context of Op HONOUR: http://roryfowlerlaw.com/an-example-of-decision-making-on-a-balance-of-probability-under-op-honour/. Specifically, I have asserted, and continue to assert, that the Administrative Reviews conducted by DMCA incorporate a boiler-plate approach to adjudication.  Although my experiences with Administrative Review represent anecdotal evidence, that evidence still has a reliable foundation.

For example, in Administrative Reviews in the context of Op HONOUR (which are increasingly common), in every ‘disclosure package’ staffed by DMCA 2, the ‘disclosure letters’ (which use largely identical text, save for the identifying information of the respondent) encloses three Annexes:

Annex A – Acknowledgement of the Disclosure Package;

Annex B – General Principle of Procedural Fairness; and,

Annex C – Assisting Officer Appointment.

All of these Annexes are identical, save for any specific identifying information relating to the respondent.  Annex A is a one-page acknowledgement form.  Annex B is a three-page description asserting DMCA’s view of procedural fairness.  Annex C is a two-page description of the role of an Assisting Officer for the respondent, even though a respondent can be assisted by a non-commissioned member or (NCM).  Every Annex C that I have encountered includes the same typo in its introductory paragraph: “1.  AO) are an advisory role in the CAF AR Process. …”

The only variation in the content of the Disclosure Package is in the “SUMMARY – ADMINISTRATIVE REVIEW (AR)” which is also enclosed under the cover of the disclosure letter.  This summary follows the same pattern.  There are six sections or parts:

1. General information;
2. Reason for Submission;
3. Summary of all relevant information;
4. Applicable Regulations;
5. Analysis of relevant information; and
6. Possible administrative actions.

The summary is then signed by an unnamed DMCA analyst.

Part 1 is a brief iteration of the respondent’s identifying information (e.g. name, rank, service number, military occupation, unit).  Part 2 is a one-sentence statement of the reason for the Administrative Review.

Part 3 represents the principal source of particularity in the Administrative Review.  It will typically list relevant events in ‘chronological’ order.[2]  It is here that the relevant facts are presented.  I elaborate below on the merits of how this information is typically presented.  The information in Part 3 tends to be presented as fact, as opposed to being presented as evidence that must be weighed by the decision-maker.

Notwithstanding the title of Part 4, most of the information presented tends to be drawn from DAOD, which are policy directives, not regulations.  The paragraphs are generally a boiler-plate iteration and summary of the same ‘relevant’ policy instruments.  For example, Administrative Review conducted for allegations of sexual misconduct will have the same seven paragraphs.  Notwithstanding that Annex B of the ‘disclosure package’ presents general principles of procedural fairness, those principles are not mentioned in Part 4.  Nor is there any discussion in Part 4 of the prevailing case law regarding procedural fairness, the weighing of evidence, or the necessity and scope of reasons.

Part 5 purports to present ‘analysis’.  Presumably, this Part applies the ‘facts’ in Part 3, to the policy regimes described in Part 4.  As I mention, supra, there is no discussion under Part 4 of the relevant common law principles, or governing case law, relating to procedural fairness, administrative adjudication, or any principles of administrative decision-making.  In fact, despite the title of Part 5, there is typically relatively little actual analysis in this Part of the ‘disclosure package’.  Here is a recent example, with information removed to protect the privacy of the persons involved (the portions removed are highlighted in bold):

In considering this case, it shall be noted that the purpose of this AR is to review the member’s conduct (not his performance). As well, the issues at stake and consequences of the decision compel the approving authority (DMCA) to provide a comprehensive decision that will be in line with the higher standard of conduct expected of CAF members by the Canadian population. The deciding authority must be convinced, based on reliable evidence that is it more likely than not that the incident occurred. The quality of the answer will depend upon the quality, the relevance, the reliability and the weight of the evidence supporting it. The information presented in the UDI report suggested that the civil standard of proof had been met and that the events as described by all the witnesses had occurred.

The CAF does not tolerate harmful and inappropriate harassment and sexual behaviour (HISB), these behaviours are inconsistent with the Profession of Arms. [Rank, Name] did not follow the direction given in OP HONOUR and demonstrated HISB when he sent numerous text messages. As indicated in the OP HONOUR order, predators and bullies who act contrary to the betterment and wellbeing of any in our ranks are neither useful in operations nor in garrison and are not welcome in the CAF.

The evidence shows that [Rank, Name] also breached DAOD 5019-5 on sexual misconduct while [location and date], [impugned action]. He also demonstrated a failure to follow the ethical principles and behaviours expected of a CAF member. Due to his social behaviour, he did not respect the dignity of all persons, did not show strength of character and he did not place the welfare of others ahead of his personal interests.

It is illustrative that, in order to redact information to protect the privacy of the respondent, all that was necessary was to remove less than one full sentence from the ‘analysis’.  This sentence was essentially the assertion of ‘facts’ described in Part 3.  The remaining text is largely identical to the text in any Op HONOUR-related Administrative Review and, as is apparent, is devoid of any analysis regarding the weighing or evaluation of evidence.

 

Weighing Evidence is Distinct from Applying Facts to Policies

The foregoing discussion highlights a distinction that appears to be lost on DMCA analysts, and which is crucial in understanding the type of decision-making for which an Administrative Review is, and is not, suited.  And Administrative Reviews pertaining to allegations of sexual misconduct offer a useful context in which to draw this distinction.

Increasingly, it appears that Administrative Reviews are being initiated following the conduct of either unit disciplinary investigations (UDI) or military police investigations (in particular, investigations by the Canadian Forces National Investigations Service (CFNIS)), but where charges are either not laid under the Code of Service Discipline, or they are laid, but subsequently withdrawn (or not preferred by Director of Military Prosecutions).  In other words, a disciplinary investigation is conducted, but the allegations and evidence gathered by the investigator(s) are not tested under the Code of Service Discipline.  And remember: not all relevant evidence is necessarily gathered in the disciplinary investigation, particularly where a suspect exercises his or her right to silence.

Specifically, the subject of the disciplinary investigation is not permitted to offer full answer and defence within the disciplinary process and before a military tribunal.

I acknowledge that this assertion is predicated largely upon anecdotal evidence derived from my own experiences as an advocate, and from what I would characterize as ‘informed speculation’.  However, in light of the dearth or relevant data and statistics offered by the CF, anecdotal evidence is the sole source of data upon which I can rely.

For example, the CF compiles and publishes a great deal of data in relation to Op HONOUR, including: the number of charges laid, the number of summary trials or courts martial involving allegations of sexual misconduct (or, as the CDS has chosen to describe the subject: Harmful Incidents of Sexualized Behaviour or HISB), the number and type of remedial measures and administrative action  taken as a result of allegations of sexual misconduct (or HISB).  However, data points that are conspicuously absent include the number of Administrative Reviews that rely upon UDI and/or Military police Investigations, but where no charges were laid or prosecuted.  That would be an illuminating statistic.

I anticipate that proponents of Administrative Reviews as adjudicative mechanisms would assert that the accused – or respondent – would have an opportunity to offer full answer and defence in the Administrative Review.

But is that truly the case?

The next article in this series will examine, in greater detail, the pitfalls of using Administrative Review as a substitute for either a disciplinary tribunal or an harassment investigation.  For the purposes of the present article, the focus is on identifying what types of analyses or adjudication appears to be done in an Administrative Review and whether the process is sufficient for each.

In effect, Administrative Reviews based upon allegations of sexual misconduct, and where there has not been a prior adjudication by, for example, a military tribunal, will (or should) conduct two distinct types of analysis.  However, based upon my experience with Administrative Reviews, it appears that DMCA staff are either unaware of this distinction, or choose to ignore it.

The two types of analysis can be characterized as:

1. The evaluation and weighing of evidence in order to determine the reliable facts; and
2. The subsequent application of these facts to the applicable policies in order to determine the appropriate course of action.

I contend that Administrative Reviews are designed to perform the latter and are markedly ill-suited to perform the former.  And it is the former where the civil burden of proof is put to the test.  Regardless of which standard of proof is applied – criminal or civil – the function of a ‘burden of proof’ is to describe the threshold that a decision-maker must apply to the evaluation and weighing of competing evidence in order to arrive at a determination of fact.  Where the decision-maker lacks the capacity to actually weigh and evaluate evidence in a fair and transparent fashion, simply asserting that it is relying on the ‘civil burden of proof’ is an insufficient justification for the lack of such capacity.

Furthermore, I contend that the very limited analysis in Administrative Reviews, particularly in the context of sexual misconduct, focuses solely on the latter, while essentially asserting the veracity of factual determinations, even when the respondent disputes the evidence against him (or her) and offers contrary evidence.

But the most significant failing of Administrative Review in terms of the first type of analysis – the evaluation and weighing of evidence – is that when purportedly applying that evidentiary standard, the DMCA analyst does not actually have all the relevant evidence to which that (or any) burden of proof is applied.  Consider the following statement, made recently by a DMCA analyst when relying upon unproven allegations from a UDI where no charges were laid and no Code of Service Discipline Process was pursued:

The information presented in the UDI report suggested that the civil standard of proof had been met and that the events as described by witnesses had occurred.

This statement was made, not under “Part 5 – Analysis of relevant information”, but under “Part 3 – Summary of relevant information”.  The DMCA analyst asserted as fact an allegation made by a complainant in a UDI.  There was no actual analysis of why that allegation was reliable.  The analyst simply asserted that it was.  There was no analysis that one would typically encounter when evidence is being weighed in the face of competing evidence.  And one reason for the absence of weighing competing evidence is that, consistent with the procedure described at article 5.3 of DAOD 5019-2, the DMCA analyst conducts this ‘analysis’ prior to disclosing the Administrative Review file (the ‘Disclosure Package’) to the respondent.

Think about that.  The analysis conducted by the DMCA 2 analyst is conducted before the respondent has an opportunity to consider the disclosure and offer his (or her) own evidence and representations.  And that is precisely the process described by DAOD 5019-2.

Assertions like the one I describe are problematic and dangerous assertions for a DMCA analyst to make when the respondent to the Administrative Review has yet to be given an opportunity to make full answer and defence.  The fact that a UDI includes allegations does not automatically mean that those allegations are reliable.  Often, the UDI will tend to present only part of the relevant evidence.  Yet the DMCA analyst asserts that the allegations must be true on a balance of probabilities and makes this assertion before the contents of the Administrative Review are disclosed to the respondent.

Undoubtedly, DMCA and her staff would take the position that the post-disclosure stage of Administrative Review represents the respondent’s opportunity to make full answer and defence.  However, the problem with that approach is that DMCA relies heavily on the ‘analysis’ of the DMCA analyst.  Where that analysis is conducted before the disclosure is made, there is a tendency for that perspective to become fixed as a rebuttable presumption.  There is a risk that decision-making by DMCA will not be conducted with an open mind.

Moreover, I contend that an Administrative Review is not the proper vehicle for such an inquiry and determination.  As I have indicated above, it is manifestly an inadequate investigative mechanism.  Administrative Review is also ill-suited to making determinations of credibility.  There is no scope for cross-examining any witnesses, including both the complainant and the respondent.  The frailty of the system is not predicated upon notions that the trier of fact must be able to “look the witness in the eyes” when the witness is providing evidence.  The shortcoming is the absence of any mechanism by which a witness’ evidence may be tested under the crucible of cross-examination.

And there are further frailties in the Administrative Review process as a fact-finding process.

Based upon my experience, over the past few years, Op HONOUR-related Administrative Reviews demonstrate consistent characteristics:

1. The ‘Summary of relevant facts’ will universally focus only on the negative facts that have been selectively gathered by the respondent’s unit and sent to DMCA;
2. The ‘analysis’ will largely be boiler-plate language that makes reference to CF policy instruments, including DAOD 5012-0, 5019-0, 5019-4, and 7023-0, extoling the obligations imposed on the respondent to act ethically, follow CF policy, and respect the dignity of others. Absent from such discussion is any recognition that the same obligations are imposed on CF decision-makers – i.e. that they are obliged to follow the rules, act ethically, and respect the dignity of the respondent;
3. There will be limited actual analysis that is unique to the specific circumstances in the Administrative Review and it will be conveyed by pejorative or value-laden language about what a very bad person the respondent is; and
4. Where remedial measures are purportedly ‘considered’ (as is typically the case in Administrative Review for sexual misconduct), there will often be no discussion of the graduated remedial approach described in DAOD 5019-4 Remedial Measures. Even if the respondent has no history of being subject to prior remedial measures (such as Initial Counselling or Recorded Warning) the conclusion will typically be that the ‘violation’ of Op HONOUR is so egregious, that only compulsory release will satisfy the objectives of DAOD 5019-2 and 5019-4.

What I find particularly remarkable in such circumstances, particularly where no Code of Service Discipline proceeding was pursued prior to the conduct of the Administrative Review, is that the alleged misconduct is evidently so egregious that it warrants compulsory release, instead of a lesser, and as-yet-unused remedial measure, but it wasn’t so egregious that it warranted a charge under the Code of Service Discipline.

That is a remarkable dissonance.  Article 3.2 of DAOD 5019-4 states that “Remedial measures are serious steps to assist a CAF member in overcoming their conduct or performance deficiency and are based on established CAF standards.”  Moreover, they are employed by the chain of command to “… (a) make the CAF member aware of any conduct or performance deficiency; (b) assist the CAF member in overcoming the deficiency; and, (c) provide the CAF member with time to correct their conduct or improve their performance.”  Where the chain of command and DMCA analysts recommend compulsory release, they have rejected using lesser remedial measures in order to assist the respondent.  Within the graduated regime established under DAOD 5019-4, that must necessarily be a conscious decision.  They are essentially concluding that the misconduct was so egregious that it warrants compulsory release as the least severe administrative response to the conduct deficiency.

Surely, misconduct that is so egregious that it warrants compulsory release would also warrant prosecution under the Code of Service Discipline.

Why would the chain of command choose not to proceed with Code of Service Discipline charges in such circumstances?

Could it be because the chain of command was aware that, if a charge were laid, the accused would often have a right to elect trial by court martial before a constitutionally independent military judge?  Could it be because the chain of command did not wish to permit the accused to offer full answer and defence in a process that was not entirely within the control of the chain of command?  Could it be that the chain of command is concerned that their overly broad characterization of what constitutes the ‘workplace’ for the purposes of DAOD 5012-0 might not withstand judicial scrutiny?

Finally, based upon my experience with Administrative Review, it appears that DMCA is largely unconcerned by the shortcomings in the application of DAOD 5019-2 in these circumstances.  A likely rationale for this lack of concern is that the decisions made by DMCA are not subject to potential judicial review by the Federal Court.  Any CF member seeking to challenge a decision by DMCA must first exhaust the CF internal grievance process.  Where the decision is for ‘compulsory release’ under QR&O article 15.01, that grievance process will take markedly longer than the ’30 days notice’ that is given for a compulsory release under items 2 and 5 of QR&O article 15.01.

 

Conclusion

Administrative Review can be an effective decision-making mechanism, particularly by centralized CF statutory decision-makers seeking to make consistent policy-led decisions.  Although some improvements can be made to the policy structure in DAOD 5019-2, it is largely consistent with the prevailing principles and standards of administrative law.  Many of the problems arising in Administrative Review arise from how the policy is implemented by those decision-makers.

Administrative Review does not constitute an effective investigative tool.  Nor does it present an effective decision-making mechanism where credibility is at stake or where competing evidence must be evaluated and weighed.  It is not a mechanism that permits a respondent to make full answer and defence to contentious allegations.  In such circumstances, Administrative Review is better suited at relying on prior adjudications by tribunals and decision-makers who are subject to legal and policy frameworks that permit such decision-making in an effective and fair manner.

Administrative Review is far better suited to applying previously established facts to a policy regime.  It is in this context that the benefits of a centrally positioned decision-maker can be realized.  While such a decision-maker lacks meaningful independence, or a capacity to weigh and evaluate competing evidence in a fair and transparent manner, the decision-maker can ensure a consistent application of CF policy.

Unfortunately, at present, it appears that Administrative Review will continue to be used for purposes for which it is ill-suited.  The issues arising from the use of Administrative Reviews as a substitute for disciplinary processes and harassment investigations and resolution under DAOD 5012-0 will be the subject of further examination in the next article in this Blog series.

(Suggested Citation: Rory Fowler, “The Nature and Limitations of Judicial Review” (June 13, 2020), online: The Military Justice Project, http://militaryjusticeproject.com/the-nature-and-limitations-of-administrative-review/).

[1] See, for example, Police Services Act, RSO 1990, c P-15, s 85; Penner v Niagara (Regional Police Services Board), 2013 SCC 19, [2013] 2 SCR 125.

[2] I use the term ‘chronological’ in inverted commas as I have, from time to time, encountered ‘Summaries of relevant facts’ that place ‘events’ in an order that does not truly reflect the chronology of a respondent’s career.  In some cases, the stated chronology disingenuously misstates the actual sequence of events.