Since there is no time limit on such warrants, it is possible that a single warrant, signed decades ago, still requires that all such ``telegrams'' be produced. Such warrants do not require judicial authorization, there is no requirement for public reporting on their existence, and no procedure exists for independent review of their necessity or appropriateness. Canadians, who are the likely participants in most of these communications, have no greater protection from this sweeping provision than do people of any other nationality.
Crossborder telephone communications probably (but not necessarily) fall under the same legal regime as domestic communications.
Non-encrypted radio-based telephone communications (RBTCs) are an exception to the above rules. The most obvious example of an RBTC is a telephone call via a cellular telephone, but it is conceivable that a telephone call transmitted by a communications satellite or even by a microwave repeater station also would qualify as an RBTC. This perhaps unlikely possibility is significant because the great majority of long-distance telephone calls made in Canada are transmitted either by satellite or by microwave repeater at some point. RBTCs are not considered ``private communications'' under the Criminal Code. The interception of such communications is legal as long as it is not done ``maliciously or for gain''. The use or disclosure of the resulting intercepts, however, remains illegal, except for certain purposes. These include use or disclosure: ``(a) in the course of or for the purpose of giving evidence in any civil or criminal proceedings or in any other proceedings in which the person may be required to give evidence on oath; (b) in the course of or for the purpose of any criminal investigation... (e) where disclosure is made to a peace officer and is intended to be in the interests of the administration of justice; or (f) where the disclosure is made to the Director of the Canadian Security Intelligence Service or to an employee of the Service for the purpose of enabling the Service to perform its duties and functions under section 12 of the Canadian Security Intelligence Service Act.'' (Section 12 authorizes the collection of information concerning ``activities that may on reasonable grounds be suspected of constituting threats to the security of Canada''. The SIRC is empowered to review such disclosures to CSIS, but only to determine whether they meet the requirements of section 12.)
It would appear, therefore, that it would be legal for CSE or any other government agency to intercept all cellular telephone calls, and all other types of ``radio-based'' telephone calls, without a warrant, as long as the subsequent use or disclosure of the intercepts remains limited to the purposes authorized by the law. Depending on what activities are interpreted as being part of ``interception'' and what are interpreted as ``use or disclosure'', it might be legal for CSE to record and examine all such calls and then ``use or disclose'' only the contents of those calls that meet the legal requirements listed above.
An even greater loophole in the legal protection of domestic communications (``private'' or otherwise) exists as a consequence of CSE's membership in the UKUSA community. This membership gives Canada extensive access to the SIGINT output of the other UKUSA member agencies, including the US National Security Agency. It is, of course, entirely legal under US law for the NSA to intercept Canadian domestic communications. It is also - apparently - entirely legal under Canadian law for CSE (and thus the Canadian government) to receive such intercepts, or any information derived therefrom, from the NSA, even if it would not have been legal for the Canadian government to intercept those communications in the first place. It appears that it would even be legal for the Canadian government to request that NSA intercept specific Canadian domestic communications. (Such an arrangement would enable the government to monitor anyone of its choosing while stating, truthfully, that no agency of the Canadian government was engaged in such activities.) It is not likely that this kind of activity currently is taking place on a large scale; NSA has more pressing targets for its eavesdropping resources. It is important to recognize, however, that there are no apparent legal impediments to such an exchange of information.
As the above examination demonstrates, one way or another it would appear to be possible for CSE to obtain intercepts of virtually any form of Canadian electronic communication without violating Canadian law. This suggests that the extent to which it does obtain such intercepts will be governed more by the government-of-the-day's intentions (as evidenced, for example, in what is known about the CSE mandate) and by CSE and allied capabilities than it will be by the legal regime pertaining to interceptions. With respect to CSE's mandate, it has been noted already that CSE is restricted to the collection of ``foreign intelligence'' (some of which does involve Canadians). It is important to recognize, however, that nothing prevents this mandate from being modified at any time, without public knowledge, if a future government should decide to use CSE's powers more systematically against Canadians. It is for this reason that the members of the Special Committee on the Review of the Canadian Security Intelligence Service Act and the Security Offences Act, and others, have called for CSE's mandate and powers to be established by statute, just as those of CSIS have been.
Aside from the CSE/SRS sites in the Ottawa area, none of these facilities is located at the logical points for the systematic interception of domestic Canadian communications. It is possible that secret facilities exist for this purpose, but since the known facilities appear to account for the entire complement of SRS intercept operators it is not likely that any sizable secret facility could be staffed by SRS personnel.
The number of CSE personnel potentially available for domestic monitoring is also severely limited. There is good reason to believe that CSE has been involved in monitoring the international communications of the Quebec government. In 1994, former CSE employee Mike Frost claimed that a ``French Problem'' section, dedicated solely to ``the question of Quebec separation,'' was operating at CSE. The government's response - that CSE does not violate Canadian law and does not ``target'' Canadians - was, in effect, a non-denial of the charge. Nevertheless, the number of CSE personnel assigned to this and any other ``domestic'' operations is likely to be small. The great majority of CSE's personnel are needed to process, analyze, and report on the agency's foreign targets.
In any case, the volume of Canadian domestic communications is far too great for any Canadian agency, or even the enormous NSA, to monitor more than a tiny fraction of it, even if Canadian communications were at the top of their priority list.
Nonetheless, some monitoring is feasible. Systematic interception of Canadian communications traffic travelling by land-line probably would require physical installations at nodal points in the communications system, but interception of the large volume of domestic traffic that travels by satellite could be done with a handful of satellite dishes at almost any point in the western hemisphere. (Russia, for example, monitors North American communications satellites from Lourdes, Cuba.) CSE probably does not monitor Canadian communications satellites, but it is a near-certainty that NSA does. It is also a near-certainty that CSE could have access to NSA's intercepts if it decided to ask for them. The crossborder traffic that is subject to section 7 of the Official Secrets Act is even easier to acquire - the Minister of Justice can order the telecommunications company carrying the traffic to provide copies directly to any Canadian government agency the minister designates.
Personnel limitations remain a fundamental constraint on the feasibility of large-scale monitoring operations, but even those limits do not rule out all forms of large-scale domestic monitoring. The initial sorting and analysis of data traffic such as e-mail and electronic financial transactions, for example, can now be done by computers using word- and topic-spotting technology; this makes it feasible to monitor a vast amount of such traffic with a comparatively small staff. Since data traffic plays an increasingly important role in the day-to-day life of Canadians, the possibility that it is, or could become, subject to systematic monitoring by Canadian or foreign intelligence agencies should be of concern to all Canadians.
Eventually it is likely that even voice traffic will become vulnerable to computer monitoring techniques. Documents released under the Access to Information Act in 1993 indicated that CSE hoped to have a topic-spotting speech recognition system ready for field testing by the end of March 1994; a fully-operational system with the additional capability of identifying individuals by their voices was scheduled to be in service by April 1995. NSA is reported to have similar technology in service already. (In fact, crude systems reportedly have been in service for a decade or more.) Speech recognition technology is not yet capable of efficient and reliable speaker-independent continuous-speech word recognition and it may not reach that point until well after 1995. But the technology continues to improve, and there is no reason to doubt that at some point it will become feasible to use computers to search systematically through voice communications just as they search through data communications now.
 There is a possibility that Section 7 also applies to crossborder telephone communications. This section was modelled on, and is essentially identical to, Section 4 of the British Official Secrets Act 1920 (which has since been overtaken by the Interception of Communications Act 1985). In 1880, a British court ruled that ``any apparatus for transmitting messages by electric signals is a telegraph''. As John Baxter has pointed out, the effect of this definition ``was to remove the possibility of any legal difference between the electromagnetic telegraph and the telephone so that a telephone message and a telegram were one and the same thing. When the word telegram was used in Section 4 of the 1920 Act it imported this absence of distinction and it does not appear to have caused any anxiety that the use of this small word was going to provide a statutory procedure for the tapping and production of international telegrams, telex messages and the transcripts of telephone calls. The Birkett Committee [established in 1957 to review the authority of the British government to intercept communications] touched upon the topic when it reported the argument put before them that the Post Office regarded the interception of telephone calls and telegrams as being on the same footing.'' (John Baxter, State Security, Privacy and Information, Harvester Wheatsheaf, 1990, pp. 186-8.)
 In Flux But Not In Crisis, report of the Special Committee on the Review of the Canadian Security Intelligence Service Act and the Security Offences Act, September 1990, pp. 152-153.
 Spyworld, pp. 96-97 and 242. It should be noted that the only examples of Quebec-related intercepts cited in the book are cross-border communications involving the Quebec government. The interception of such communications would seem to indicate a willingness on the part of the Canadian government to use CSE to collect intelligence related to Canadians and to domestic Canadian political issues, but it does not imply necessarily that CSE intercepts domestic communications for this purpose. Frost states on p. 96 that ``I can't say I ever saw anybody actually doing Quebec intercept from CSE or any raw copy coming from there.'' Similar allegations regarding communications between Quebec and France have been made in the past: see, e.g., Graham Fraser and Madeleine Drohan, ``Embassy officials ridicule Canada-France spy story,'' Globe and Mail, 22 May 1992, p. A5.
 See, e.g., House of Commons Debates, 25 October 1994, pp. 7178-7179. As noted in the section on the legality of monitoring, an assurance that CSE is not breaking the law is not an assurance that communications involving Canadians are not being intercepted. An assurance that CSE is ``not authorized to target Canadians'' is equally unreassuring. CSE's targets are not normally individuals but, rather, various categories of ``foreign'' intelligence, such as, in this example, the nature of discussions between the governments of France and Quebec. An assurance that no Canadian is targeted is not an assurance that no communications involving Canadians will be deliberately intercepted. For a description of NSA's use of the same semantic dodge, see James Bamford, The Puzzle Palace, Penguin, 1983, pp. 387-388.
 See contract W2213-3-1903, Annex B (``Statement of Work For the Research and Development of A Topic Spotting System,'' March 1993), p. 11, released in severed form under the Access to Information Act.
 ``Key words trigger U.S. security system,'' Ottawa Citizen, 29 July 1994, p. A9. Spyworld, p. 152, claims that CSE was given access to a highly effective NSA wordspotting system, codenamed ``Oratory,'' as early as 1982. A reliable, but dated, description of speech recognition efforts by the UKUSA SIGINT agencies can be found in Patrick Fitzgerald and Mark Leopold, Stranger on the Line: The Secret History of Phone Tapping, The Bodley Head, 1987, pp. 104-111.
To be contacted for a confidential consultation |
please E-mail: firstname.lastname@example.org
or send a letter via US Mail to: