Report 20 - Access to Information Review Task Force

THE ACCESS TO INFORMATION ACT AND THE SECURITY AND INTELLIGENCE COMMUNITY IN CANADA

I. Historic Records

"Historic records" are defined for the purpose of this paper as records of historical significance that have been preserved in readable form and have reached an age whereby they are no longer of conceivable, current operational significance. I have not attempted to put an age to such records, but in my view 30 years could serve as a rough, universal benchmark. A 30-year benchmark would bring S&I records into alignment with the public release of Cabinet documents.

Historic records should be seen as forming an important part of the bargaining equation that requires the security and intelligence community to balance the need for public knowledge against the requirement for maintaining security. On the face of it, historic records would seem to offer the least burdensome problem to the security and intelligence community and the area where the community could most easily provide public assurance of its fidelity to the information bargain through release of materials.

Yet, this has not happened. Canada lags far, far behind its principal intelligence allies, the United Kingdom and the United States, in terms of the amount and quality of historic records released. There is no sign of any effort to address this disparity, which given the far more sensitive holdings enjoyed by both our Allied powers, is paradoxical to say the least. It may even be the case that the immensity of the disparity is not understood.

The "problem" of historic records is thus that they call into question the effective functioning of the Access "bargain."

Before discussing the problem of historic records in more detail, three inter-locking points need to be made. The first concerns what we might call the philosophy of the Access system; the second concerns a rather naïve assumption built into the original Access legislation; the third concerns the absence of any time limits attached to exclusions under the Act.

Whatever was the intention of the framers of the Access to Information Act Legislation in the early 1980s, it must be recognised that it is a system designed to serve a vision of "consumer demand." Access was to meet a putative citizen demand for information about government operations by providing a system whereby individuals would have the right to request records and would be presented with a relatively transparent mechanism for monitoring the progress of their requests and for disputing outcomes perceived as unjust. As a system, Access was based on a mistaken metaphor. Citizens might be consumers, but government records are not a consumer product. The net effect of this approach was to ignore the value of government records as a coherent, intricately related and frequently indivisible body of documents in favour of an atomistic approach. Such an approach, in effect, transformed an "archive" into a shopping mall, with individual consumers wheeling their baskets around in search of individual products. Add to the strangeness of this notion, the fact that Access consumers often had to wheel shopping baskets around with a blindfold on.

It is hard to conceive of a metaphor translated into practice that is more likely to be injurious to a real historical understanding of the significant dimensions of Canadian history, in the security and intelligence field as in many others.

It can, of course, be argued that the Access legislation was not meant to provide the only form of public release of historic government records. Indeed, this argument was explicit in the Access Act. The Act (R.S. 1985, c. A-1, s.2) reads:

"This Act is intended to complement and not replace existing procedures for access to government information and is not intended to limit in any way access to the type of government information that is normally available to the general public."

But this has proved to be a naïve hope, especially in regards to security and intelligence records. There are currently no other mechanisms in place for the declassification and release of historic records relating to security and intelligence other than Access, and no identifiable will to open up other mechanisms. ⁽³⁾ The result is an empty shopping mall, with only a tiny handful of persistent researchers. The scene is uncomfortably like the experience of shopping in the Soviet Union in its bleakest eras, Moscow's G.U.M. on a bad day.

To add to the problem posed by historic records is the fact that there are no time limits specified in any of the exemptions. Thus records may be exempted, in theory, for "eternity", a patently preposterous idea. In practice, they can be exempted to a point in time when their relevance can no longer be easily judged, their context no longer framed. Locked away in no one's memory, their state of preservation and readability of little matter, such records can easily "fade" away, both figuratively and literally.

The current outcome of the Access Act as it relates to historic security and intelligence records is that it has failed to provide a workable system for the release of coherent and significant blocks of records, the only kind of records from which history can be adequately written. Instead, we have an atomised system in which Access allows for bite-sized historical scholarship to exist, but effectively prevents anything more synoptic. The history of the Canadian security and intelligence community is thus little understood. The obstacles to scholarship in this area are formidable, starting with an absence of foundational knowledge and proceeding to a perception of the Access mechanism not as an aid to scholarship, but, because of the time involved, the uncertain outcome, and sometime the costs involved, as a real deterrent. The absence of open historical records means that our understanding of the history of Canadian endeavours in security and intelligence is susceptible to being shaped from offshore records, particularly British and American. Such a situation holds out the obvious dangers of distortion and misperception.

The question that needs to be asked is: if it was not the intention of the Access Act to monopolise the processes for records releases, why has this occurred in the security and intelligence field?

The answer to this question lies, I believe, in three areas: resources, will, and bureaucratic logjams.

To put the resource issue in context it must be understood that the security and intelligence community has, as a first call on its resources, its own information operations. The S&I community operates a classic intelligence "cycle," whereby it must, as its life-blood, collect, analyse, and disseminate useful information to government decision-makers. This is a demanding function, made more so in the current S&I environment by the complexity of issues that it must face in a world no longer dominated by the familiar dynamics of the Cold War, and by the cascading flows of information opened up by the communications revolution. Historically, the Canadian security and intelligence community has always struggled against problems of resource scarcity, underfunding, and a lack of understanding on the part of decision-makers. Arguably, the situation persists today.

The S&I community has a job to get on with; it does not have extra resources to devote to a proactive exercise in document release. By this I mean a process whereby an agency in the S&I community would decide to survey in depth, sever where necessary, and open up its historic records for public use in the National Archives. This point was made to me in conversation with Tom Bradley, Assistant Director and head of the Canadian Security Intelligence Service (CSIS) Secretariat. CSIS is obviously proud of its Access unit. It devotes considerable resources to Access, with a section comprising at present 13 officials. In my consultations I found that CSIS, unlike many other government agencies, often gets high marks for its professional handling of Access requests. ⁽⁴⁾ Such requests have increased considerably since 1992 and CSIS does not feel it has resources to add to deal with proactive releases. Yet, in the case of historic records under CSIS control, only CSIS could act in a proactive manner. Because of a mandatory consultation protocol with the National Archives, only CSIS can take the lead in a proactive strategy.

Resource issues intersect with questions of will and desire. From my consultations with officials at CSIS, the Communications Security Establishment (CSE), and the Privy Council office, it seems clear that there is little will or interest in taking a lead in opening up historic records. In my experience, this is not because the community lacks an interest in its history or is unaware of the perils of loss of historic memory. Rather it is simply a question of priorities. A small and resource-stretched intelligence community, even though it has a rich history, is inevitably consumed by its focus on current operations, current problems, and day to day business. Moreover, the security and intelligence community is given little incentive to think about historical documents as a separate and special category of records. Not only does this cut across its own traditional sensitivities about secrets, and fear of precedent-shaping actions, but the timelessness of intelligence records is explicitly sanctioned by the nature of the exemptions provided under the Access Act.

A good expression of the prevailing wisdom on the issue in the S&I community was contained in a presentation given by a senior Access official at CSIS, Normand Sirois, in an address to the Canadian Historical Association annual meeting in Edmonton, Alberta in May 2000. M. Sirois stated:

"Some of you might argue that documents of 50 years and older should be exempt from exemptions because of the passing of time. Our experience at CSIS says otherwise. We are of the opinion that there is still a risk in disclosing old documents. Public knowledge of the degree to which certain methods and procedures are/were used reflects not only the level of the threat posed by particular subjects of an investigation, but also the extent of the Service's knowledge and interest in their activities. With such knowledge those whose activities are inimical to Canadian interests could assess the depth and sophistication of our resources, as well as the degree of expertise of the Service. Through analysis, these individuals and groups could predict the Service's endeavours, and thereby counter any potentially effective investigative thrust." ⁽⁵⁾

Put in historical context, M. Sirois is talking here of records generated by the RCMP Security Service no more recently than 1950, at which point the Security Service, which no longer exists, was engaged principally in counter-intelligence efforts against the Soviet Union, a state that no longer exists, and in monitoring of dissident and other political movements deemed subversive, few of which exist today. In my opinion, M. Sirois's argument is difficult to credit. But I cite it not to engage in debate, but rather to underline the point that CSIS and other agencies in the security and intelligence community see in the issue of the release of historic records potential dangers, resource costs, and no obvious gain relative to their quotidian business.

While it might be argued that such an attitude betrays a lack of enlightenment about the value of history as a tool which might be beneficial in building institutional morale, in internal training, or in securing political and public acceptance, the fact is that enlightenment comes through experience. There is a "Catch 22" at work in regard to historical records. The value of their release is hard to demonstrate when they have never been released.

This leads to consideration of a third aspect of the historic records problem--that which I have labelled bureaucratic logjam. Even if an outlook existed in the S&I community in favour of the release of historic records, there is no effective leadership structure in place to allow such a thing to happen. The Canadian security and intelligence community is itself a decentralised system. Different, specialised intelligence tasks are parcelled out among a number of agencies, with only loose co-ordination from the centre, as exercised by the committee and secretariat structures managed by the Privy Council Office, ultimately culminating in the office of the Co-ordinator, Security and Intelligence. The structure of the community places the onus for records release strategies on the individual agencies. The individual agencies do not themselves perceive any advantage in the voluntary, proactive release of historic records.

There are, of course, other players in the Access system. There are the professional archives staff at the National Archives of Canada. ⁽⁶⁾ I met with senior archivists to discuss this issue. The mandate of the National Archives involves them in negotiations with government departments to arrive at retention schedules for documents, which govern the process whereby government records are distinguished as suitable for preservation or destruction, are transferred if preserved from the control of departments to the National Archives, and are eventually deemed to be "historical" records ready for public release. The principal interest of the archivists is in ensuring the orderly retention, transfer, cataloguing and preservation of records. They have, of necessity, to take the long view and it was expressed directly to me that the National Archives as an institution cannot afford to be seen to be out "ahead" of government departments in terms of records release strategies. Such a thing would endanger close working relationships with the government departments and perhaps the fate of records themselves. Like the intelligence community, the National Archives see their job in terms which preclude them from taking a lead in a proactive strategy for the release of historic records.

The Access Division of the National Archives (organizationally separate from the rest of the Archives staff) faces a similar constraint. ⁽⁷⁾ Sarah Gawman, the Acting Director of the ATIP Division at the National Archives, was very helpful in taking me through the various options that exist for the release of government records. She identified three processes that can be applied. The first is formal requests via Access for government documents. Such formal requests, in the case of CSIS records, always are sent to CSIS's ATIP shop for review, under the agreement that exists between the Archives and CSIS. In the case of records from other government agencies in the S&I sector, they sometimes go to the departments for review, as is often the case for DFAIT records, and always involve a process of consultation back and forth between the Archives and the Departments. Sarah Gawman indicated that close working relations between the Access division at the National Archives and departments allows the Division to be in a position to establish likely requirements for time extensions to the initial 30 day reporting requirement on Access requests. She particularly mentioned the effective working relationship with CSIS Access in this regard. Extensions beyond the 30-day period seem to be near universal and are, in part, a consequence of the requirement for consultations. I understand from conversations with Sarah Gawman that the average extension period is 90-120 days, three to four months.

A second process Ms. Gawman described as "informal review." Here, the Access division, in responding to a request for records from a researcher, agrees with the researcher to engage in an informal review of material outside the Access procedures. This allows the Access division to handle some complicated research requests that might be part of a more long-term research project. For its part the Access division, in handling informal reviews, is not constrained by the formal 30 day reporting requirement, or subsequent extensions. The researcher, for her part, is not required to pay any fees, but cannot use the complaint process should materials not be released. Of course, the researcher can decide, at a later date, to switch tracks from an informal review to a formal Access request and so have recourse to the complaints process. But should a researcher find himself or herself in such a position, the cumulative delays involved would be great and the experience less than satisfactory.

Both the formal and informal Access review paths are, of course, purely consumer driven. In my view, neither process is likely to lead to the release of records that are sufficiently comprehensive to allow for the encouragement of broad-based research and the generation of public knowledge.

A third process involves what is called a "block review" of records. This process is essentially archives-driven, although the Archives might take into account expressions of researcher interest in certain types of records. However, there is no systematic means of establishing public interest in records or of taking expert advice from the academic community. There are a number of criteria established for the block release process. One is that the records in question have to be over 20 years old. Probably most importantly, the records must be deemed to be of low sensitivity. If these criteria seem to hold, the Access section, in consultation with archives staff, sample a block of records for a targeted class--checking something on the order of 5% of the records involved. If the sampling process confirms that the records are unlikely to contain anything sensitive, then it would be determined that such a category of documents could be released in their entirety without having to be processed on a page by page, file by file basis. This process has been applied to files from such areas as Agriculture, Parks Canada, Public Works, Lighthouse files.

As can be readily seen, the block review process, though it has potential application to historic S&I materials, would, in view of the existing criteria, never be applied to such records. Sarah Gawman confirmed as much to me, as did Bob McIntosh, the Chief of the State, Military and International Affairs division of the National Archives. Yet the United States intelligence community is able to apply a block review process to old intelligence records, suggesting that it can indeed be done, despite the negative outlook in Canada. We could learn from the American experience in this area.

The conclusion I draw is that in the area of historic security and intelligence records we face a stalemate. The stalemate is frozen in place by a number of factors:

• the lack of time limits for exemptions under the Act
• the lack of resources to engage in proactive release of materials
• the lack of will on the part of those who control such records to undertake proactive release
• the absence of centralization and leadership on records issues in the Canadian security and intelligence community
• the inability of the National Archives to take a lead.

The outcome of this stalemate is that the explicit intention of the Access Act not to interfere with other processes for the release of records cannot be met. Instead the entire process for records releases is controlled by the Access Act. The result, in my view is a very fragmented and unsatisfactory system, driven by "consumer demand." This system can only result in the effective destruction of a coherent "archive" of government records and its substitution by an atomised and essentially random collection of documents the initiative for release of which depends entirely on individual researchers, themselves usually working in isolation.

Conclusion and Recommendations re: Historic Records

The stalemate, which I have described, is insupportable and must be broken. A system must be put in place which allows for the coherent declassification and release of records of historical significance in the security and intelligence area that is not dependent upon the Access process. Until such a system is in place, it is my view that the essential bargain implicit in the Access Act is not being met and that the trade-off between the powers of protecting sensitive documents and the need to provide for openness is being abused.

To achieve a coherent and rational release strategy for historic records, I would recommend for consideration the following measures:

1. The establishment of time limits on exemptions under the Act. The time limits might have different thresholds depending on the sensitivity of the records, starting with 20-30 years, and allowing for increases to 50 and 75 years. But such extensions would need to be approached with fine judgement before they were applied, their application should be limited according to a serious injury test, and should be subject to periodic review.

2. Greater transparency with regard to schedules of document transfers from Departments to the National Archives.

3. The compilation and publication of a National Archive guide to historic security and intelligence records.

4. Encouragement and resources to be given to departments in the S&I community to identify and process in "block review" form historic records to be opened and transferred to the National Archives.

5. An effective start be made immediately on the release of aged security and intelligence records. This could be done either by a process of block review according to specified time period, as per the NATO records strategy, or by starting with a significant, central collection of records such as the minutes and papers of the Joint Intelligence Committee and the Joint Intelligence Staff for the period 1943-1970.

6. Recognition of the need for central leadership and responsibility for historic records release policy, to be vested in the machinery of the S&I secretariat of the Privy Council office and to be part of the mandate of the Co-ordinator, Security and Intelligence. Additional resources would be required to allow the Co-ordinator to deal with this work.

7. The formation of an advisory panel on intelligence community historical records, to consist of government officials and academic experts, who would meet on a regular basis to identify the sorts of historic records whose release should have priority and who would work with the intelligence community to achieve such releases. ⁽⁸⁾

8. The creation of a small historical staff to serve the entire security and intelligence community. The mandate of this staff, along the lines of the CIA's Centre for the Study of Intelligence, would be to promote historical knowledge throughout the community, to assist in training, and to serve as an outlet for the publication of historical reports.

II. Contemporary Records

For the purpose of this discussion, I will use the term "contemporary records" to distinguish between historic records and materials of more recent vintage. As a rough benchmark, I am referring to security and intelligence records younger than 30 years, which are probably held by departments rather than being under the control of the National Archives, and which may have on-going operational or policy relevance. In my view, given the fast-changing nature of the international and domestic security environment, and the focus on current events and operations that is a feature of the Canadian community, it is unlikely that many records held are of genuine operational significance beyond a five to ten year time span.

With regard to contemporary records, we enter a different spectrum of the bargain discussed at the outset of this paper. In my view, the balance of the Access bargain must shift, from a predilection to openness (historic records) to an emphasis on the protection of secrets and due regard to sensitivities.

But there are problems involved in making the shift. One is that given the failure to operate the Access bargain successfully with regard to historic records, the security and intelligence community, by failing to distinguish between historic and contemporary records, is under increased pressure to apply a uniform standard to them all. Expectations, so this argument would go, for the release of contemporary records are therefore higher than they might be were a larger body of historic records available for study and comprehension.

A second problem, not unrelated, is that given the generally low levels of public understanding about the security and intelligence function in Canada, there is a general disbelief in the notion that Canada has secrets to protect. Scepticism about secrets places an additional and unnecessary burden on the security and intelligence community to be seen as acting scrupulously and fairly in regard to the letter and spirit of the Access Act.

The Access Act proclaims a "right of access to information in records under the control of a government institution in accordance with the principles that government information should be available to the public, that necessary exceptions to the right of access should be limited and specific and that decisions on disclosure of government information should be reviewed independently of government." ⁽⁹⁾ In my view, this vision of a right of access to contemporary records in the security and intelligence domain is not really tenable. Such a right of access, if taken literally, would be unworkable and would threaten the ability of the community to do its job and to maintain its fundamental working relationship with its allied partners. While this may seem a surprising thing for an academic researcher to say, in fact the principle that the security and intelligence community has secrets and a stringent requirement to protect them is universally accepted among serious students of the community. The sooner we face this fact and its implications, the better.

What we thus have is an Access Act that operates, of necessity, as a fiction in regard to contemporary records. Public demands under Access can be countered by the application of the major clauses of exemptions, both mandatory and discretionary, offered under the Act. In the Security and Intelligence realm, the principal exemptions of relevance are Section 13 (Information obtained in confidence), Section 15 (International Affairs and Defence), Section 16 (Investigations, threats to the security of Canada) Section 21 (Advice and Recommendations). Altogether, the exemptions are a formidable defensive mechanism in the hands of the community to protect secrets. Both the Canadian Security and Intelligence Service and the Communications Security Establishment, the two main collectors of sensitive intelligence in the community, regard the Access Act as offering sufficient protection. But servicing the Access Act can place a serious demand on the resources of the security and intelligence community, and face them with trade-offs between Access work and intelligence collection, analysis, and reporting capabilities. Consider, for example, that the size of the Access section at CSIS is roughly equivalent to the strength of the Intelligence Assessment Secretariat of the Privy Council, responsible for strategic foreign intelligence assessment and co-ordination of intelligence analysis across government departments.

If the Access Act is a fiction in regard to contemporary S&I records, as I have argued, the question that needs to be addressed is whether it is a necessary fiction?

There are three possible kinds of responses to this question:

One argument would have it that the security and intelligence community is over-protected under the Access Act, that it is over-zealous in its desire to protect information, and that it should not be given such wide-ranging exemptions under the Act. That, in essence, the Access Act fiction is unnecessary and should be dismantled in order to place the security and intelligence community alongside other functions of government, with no special status as regards its records.

A second argument is that the fiction is unnecessary because it serves no public good. It wastes government resources that could be fruitfully applied in other dimensions, including strengthening the operational capabilities of the security and intelligence community itself. It also creates a public frustration as Access requesters discover, individually and cumulatively, how little material the government is willing to divulge. Thus, the argument goes, the Access Act should be discarded altogether in the security and intelligence domain. The Security and Intelligence community should be exempted in its entirety from Access requests to contemporary records and should be allowed to reorient its resources towards its operational requirements.

Both of these arguments are, of course, straw men, designed in their implausibility, to force us to think more clearly about the remaining answer to the question of the fictional nature of Access. Neither a complete absence of right of access to contemporary intelligence records, nor a situation weighted in favour of complete openness has any merit or any chance of existing in the real world. Compared to the extremes, a compromise position looks better and better.

The third argument is that to whatever degree Access operates as a fiction, promising the right of public access but simultaneously denying it, that nevertheless this fiction is necessary and can even be productive in the realm of contemporary records. It is also capable, I would argue, of being fine-tuned, as experience is gained and lessons learned. In this regard, we have to remember that the Access legislation is relatively recent, and the acceptance of the idea of the utility of high levels of public knowledge about security and intelligence activities more recent still.

The idea of the Access Act as a necessary fiction may not sound very palatable. But perhaps in practice it can be made to work. In the first place, it recognises the fact that there is an inevitable friction between the desire on the part of an intelligence community to protect secrets and a desire on the part of the public for knowledge about an important dimension of government operations, especially when those operations, for better or worse, are deemed, thanks to popular mythology, "sexy." There can be an educative outcome in the tug of war between those who would protect secrets and those who want access to them. Faced by demands for access over a period of years, the security and intelligence community has to become accustomed to exercising good judgement about what records can be released and what records cannot. The community also might grow accustomed to the idea that public knowledge does not necessarily spell harm or disaster. The public, for its part, gains, in the tug of war, some limited knowledge about the security and intelligence community and perhaps a more realistic appreciation of the gravity of the secret problem. Ideally, experience of the access function tends towards a more liberal approach to the release of government records by the security and intelligence community as caution and fear are overtaken by routine and reassurance about the consequences of records releases.

The necessary fiction of Access is also made supportable to the extent that it helps contribute to a practice whereby government departments feel able to take the lead and issue, outside the Access process, a stream of public documents about their work. One example would be the decision to issue an annual public report by the Director of the Canadian Security Intelligence Service.

The Access process is also supported by the variety of accountability mechanisms already in place to monitor the intelligence community and issue public reports. The Security Intelligence Review Committee and the CSE Commissioner, through their annual reports and other releases add to the fund of public knowledge about security and intelligence issues. The Inspector General's office in the Solicitor General's department also has a function in this area, though it tends to produce relatively little material for the public domain. The Auditor General's office has also released some material of relevance, following on their studies of the intelligence sector.

In this and other ways, contemporary records reflecting on the activities of the S&I community make their appearance in the public domain. It is a highly fragmented, limited, and controlled process for public access, but it is better than nothing.

The question is: can the necessary fiction of access to contemporary records be improved on? Or, to put it another way, can the bargain be better struck? My view is that it can.

The solution involves action in three areas. The first is a simple exhortation to the security and intelligence community to accept the need for reasonable levels of public knowledge about their activities in order to make for good public policy. To a certain extent, acceptance of this need requires a change in thinking and culture, but it is one that I believe is already under way. However, a shift in thinking, to be sustained, does need fostering by senior management and should be incorporated in the processes of recruitment and training in the S&I sector. The communications function in the security and intelligence community needs to be recognised as important, be given support and a reasonable degree of independence. Both CSIS and CSE have made important strides in this area, which should be continued and expanded upon. The temptation to use public releases as a form of propaganda or special pleading must be avoided at all times.

A second area of possible improvement involves leadership from the centre in terms of establishing support for public communication, co-ordinating the flow of information, and establishing a central node of publicly released information which can be easily utilised. Such a strategy would involve the Co-ordinator, Security and Intelligence, in the Privy Council Office in establishing and monitoring policy, and might take the form of the creation of a central communications office for all matters to do with security and intelligence across government departments, to be run from the S&I secretariat of the PCO. A degree of centralization should also be accompanied by the creation of adequate public information tools, especially the creation of a web site that would provide for links to individual department and agency sites and would also serve as a place for one-stop shopping for both currently released information and an archive of electronic texts of material previously released. As part of this strategy, the Co-ordinator, Security and Intelligence, should be required to release an annual public report on the activities of the sector.

A third area of improvement involves proactive efforts on the part of the security and intelligence community to reach out to key public sectors. In recent years, the community has shown strong support for and interest in the work of the Canadian Association for Security and Intelligence Studies (CASIS). The community has helped fund annual public conferences by CASIS and some community officials have taken part in these conferences and related fora. In my view, continuing support for academic endeavours such as CASIS, whose mandate includes an effort to increase public awareness of security and intelligence issues and to promote scholarship in the area, is crucial.

Current community strategies with regard to contact with the journalistic and media community are less easy to discern. The relationship between the media and the security and intelligence community will always be fraught with tension and suspicion, the price to pay for an independent fifth estate. But the security and intelligence community might go farther than present practice, so far as I understand it, by considering hosting an annual meeting with interested media officials, or some system of occasional, co-ordinated briefings.

My personal view is that the nature of security and intelligence practices in Canada is only fitfully understood by the media. There are some strong exceptions to this rule, especially in print journalism, where a handful of Canadian reporters have developed a considerable expertise in the area. But the price of ignorance needs to be faced, and can be high. While academic scholarship might have a long-term impact on shaping public understanding of the community, the immediate and most visceral shaping is done by the media. Anything that can be accomplished in a non-manipulative manner to increase media knowledge about the security and intelligence sector can only help provide context and ameliorate an inevitable tendency on the part of the media to seek out isolated and ephemeral stories of failure, institutional breakdown, and scandal.

There are many other ways in which steps might be taken to lower the Chinese wall that exists between the security and intelligence community and the public. These need not be detailed here, but their general thrust needs to be understood. Creating reasonable levels of public knowledge about security and intelligence matters, through the release of contemporary records, is not just a matter of accountability. It is, more fundamentally, a matter of generating public intelligence about intelligence, which in the long run is vital for the success of the S&I sector. Without public intelligence, the acceptance and understanding of the security and intelligence community's work will remain low, recruiting new generations of talent will remain difficult, the impact of the community within government will be hampered, and the general efficiency of the community will be reduced.

The Canadian security and intelligence community, historically, and one must presume, contemporaneously, has always suffered from low levels of resources and insufficient attention to their work in policy-making circles in government. Historically, its analytical capacity has been limited. The only way to escape from this dilemma is to foster public intelligence, and to sow the seeds of knowledge. In particular, in my view, a new capacity for systematic outreach on the part of the community to experts in academe and the public sector, including the business community, is vital in order to bridge the gap between likely levels of internal analytical capacity and the requirements of understanding complex changes to the domestic and international security environment. In short, information release strategies are not about public relations; they are about knowledge generation and a two-way tapping of expertise.

To come back to the specific issue of the Access Act and contemporary security and intelligence records, we face a situation in which the only reasonable approach is to live with severe limitations on the public "right" to information in this sector. Information released through the Access process will remain meagre and highly fragmentary. The requirement to protect secrets, the intelligence trinity of sources, targets, and methods of operation, to which might be added highly sensitive details about allied relationships, and the contents of most analytical documents, will and must have the upper hand. The "bargain" over secrets versus open information will be heavily weighted in favour of secrets. But as a check on the tendency of all security and intelligence communities, not just the one in Canada, to be overly protective of information and to enshrine the doctrine and even mystique of secrecy, there are alternative paths of information release, whose success depends on the intelligence community understanding that it is in their best interest to support reasonable levels of public knowledge and to have the means to draw on expertise in the public sector in intelligent ways.

To the argument that this task, the generation of public knowledge, is not in the job description of the security and intelligence community, the answer can only be that it might not have been in the past, but must be in the present and future. Moreover, there are no signs on the horizon that the burden of public knowledge generation can be off-loaded elsewhere. The Access Act effectively atomises public information. The academic community cannot take the lead single-handedly and, in Canada, is generally indifferent to security and intelligence issues. There is no sign that a Canadian equivalent of the Washington-based National Security Archive, an unique endeavour that bridges scholarship and commercial enterprise through use of the Freedom of Information Act to gather vast stores of government secrets, is ever likely to emerge. ⁽¹⁰⁾

Recommendations with regard to Access and Contemporary Records:

9. The Security and Intelligence Community must continue to have at its disposal the power to apply exemptions under the Access Act to protect information whose divulgence would be harmful to national security and the conduct of international affairs. The current Access exemptions provide powerful and sufficient tools to allow for such protection.

10. The different agencies of the security and intelligence community should continue and expand their programmes, outside the framework of Access, for the release of contemporary records under a proactive strategy.

11. The communications function of agencies in the security and intelligence community should be strengthened.

12. There should be increased centralised leadership and centralised policy-direction for information release matters provided by the Co-ordinator, Security and Intelligence, in the Privy Council Office.

13. To overcome the atomisation of information that is a feature of Access, there should be a centralised node for information available to the public on the security and intelligence community. This node, primarily in the form of a web site, would complement existing agency websites and offer links to them.

14. There should be a consolidated database of publicly available contemporary records dealing with the security and intelligence community. This consolidated database would contain a community wide set of all public releases and Access documents plus an electronic database of archival material reflecting previous releases of information.

15. Efforts to form strategic partnerships with academic and private sector groups and individuals knowledgeable about security and intelligence issues should continue and be strengthened. These strategic partnerships would allow for a two-exchange exchange of knowledge on pertinent issues.

Return to Table of Contents

Next Page

³ It might be relevant to note two programs that seem to offer exceptions, but, in fact, do not. One is a small program mounted by DFAIT that allows selected scholars access on a controlled basis to closed records. This program does not, however, apply to research in the areas of security and intelligence. DND maintains as part of its Directorate of History an archive of military records available for consultation. But this archive does not contain any significant holdings of intelligence records. Most of the material held relates to the two world wars.

⁴ My respondents variously ranked departments in terms of their handling of Access requests. CSIS often came out near the top; DND and DFAIT were often listed at the bottom in terms of researcher satisfaction; PCO had a mixed track-record.

⁵ Address by M. Normand Sirois to a panel on Access issues, Canadian Historical Association meeting, Edmonton, Alberta, May 2000. Copy made available to me by M. Sirois.

⁶ There are currently nine archivists working at the National Archives in the division responsible for "State, Military and International Affairs records" which covers most of the agencies in the S & I community.

⁷ The Access Division of the National Archives also clearly suffers from personnel shortages. The Division currently employs 19-20 FTEs out of a total authorized strength of 28.5 FTEs.

⁸ The CIA established an external review panel to consider the issue of the incorporation of intelligence community records into the published series of State department documents in Foreign Relations of the United States, a principal tool for researchers. The experience of this panel was not altogether successful and lessons should be learned from it in creating a similar body in Canada.

⁹ Access to Information Act R.S. 1985, c. A1, s.1

¹⁰ See the first annual report of the National Security Archive, posted at www.nsarchive.org, July 2001.