Photo of CSE's Cray X-MP/11

The photo above shows the Cray X-MP/11 (later upgraded to X-MP/14) supercomputer that CSE operated from 1985 to 1995. 

The X-MP/11 was CSE's first supercomputer, purchased as part of a concerted effort to revitalize the agency's limited cryptanalytic capabilities. 

By its own estimation, CSE now has "the largest concentration of supercomputers in Canada," but prior to the purchase of the X-MP/11, the computer resources available for cryptanalysis at CSE were mostly limited to PDP-8 and -11 minicomputers and occasional use of an IBM mainframe. 

I wrote about CSE's acquisition of this computer more than thirty years ago. As you can see, many details of the computer and its purchase and subsequent upgrades were known even back then. 

But no photos of the machine seemed to be available. The best that I could find in recent years were some pictures of circuit boards from the computer (along with some from a Cray-1S originally owned by Environment Canada) that are now held in the collection of Ingenium, the organization that runs the Museum of Science and Technology.

Instead, you had to make do with the schematic diagram on the right, which I obtained through the Access to Information Act and included in my 1992 article.

As it turns out, however, there has been a photo of the X-MP/11 circulating in the wild since 2018. That's when CSE partnered with Ingenium to create Cipher | Decipher, an "interactive exhibition exploring the past and present of communications cryptology."

The exhibition has been travelling around Canada ever since. It is currently on show in Strathroy, Ontario, near London, and last month I made a special trip to check it out. As CSE notes, the exhibition is "specifically designed for school-age children," using puzzles and interactive displays to spark interest in cryptology and the cyber security field more generally. But it also makes an interesting half-hour or so visit for adults.

And, for me at least, one of the most interesting and unexpected aspects was that one of the displays contained a photo of the never-before-seen X-MP/11. 

(Who would have guessed it was decorated in the colours of a 1970s kitchen appliance?)

The caption accompanying the photo was also pretty interesting:

In the 1980s, the CSE acquired this computer to exploit encrypted communications. In the later years of the Cold War, Canada and its allies analyzed Soviet ciphers with advanced computer systems. These systems were able to test possible keys, looking for the one that would unlock the cipher.

There is good reason to suspect that CSE's Cray was also used for non-Soviet targets. 

But it's no surprise that the display doesn't talk about that. Given CSE's track record on transparency, I'm actually a bit surprised they were willing to (more or less) confirm the X-MP/11's use against Soviet targets. 

Billion dollar budget

The Main Estimates for Fiscal Year 2024-25, tabled in parliament on February 29th, reveal that CSE is on track to receive its first official billion-dollar budget: $1,041,683,002 to be precise ($1,061,532,258 if you include $19,849,256 in projected revenue). 

The agency may actually cross that threshold during the current fiscal year, 2023-24. CSE's projected spending in the FY 2023-24 Main Estimates was a mere $965,909,359, but increases in its budget authority during the course of the year boosted that total to $1,039,192,674. That said, it is normal for a portion of the agency's budget authority to go unexpended during the year, so CSE's actual 2023-24 spending could very well fall short of the billion-dollar threshold.

In the FY 2001-02 Main Estimates the agency's projected budget was $100.2 million, or about $170 million in today's money.

In other words, CSE's 2024-25 budget is projected to be TEN times as large as its pre-9/11 budget in nominal terms, and even after adjusting for the effects of inflation it will be SIX times as large.  

Six times!

[Update 10 March 2024: The cyber security (formerly information technology security) side of the agency has grown the most. The cyber security side now accounts for about one-third of CSE's budget, while it was more like 23% in the pre-9/11 days. This means the cyber security budget is about 8.9 times its pre-9/11 size, whereas the SIGINT (and now cyber operations) budget is about 5.3 times its earlier size, after adjusting for inflation.]

Interestingly, CSE's staff is only around 3.5 to 4 times as large as it was in 2001. There are probably multiple reasons for the difference: a higher proportion of the budget likely goes to IT systems and services and to agency facilities; pay and benefits packages have probably improved; and the agency may also be contracting more work out to the private sector.

The trajectory of the Canadian Security Intelligence Service's budget makes for an interesting comparison. CSIS has also seen a lot of growth since 9/11, but not on the same scale as CSE. The projected CSIS budget in FY 2001-02 was $170.4 million, or about $290 million in 2024 dollars. CSIS’s projected budget for FY 2024-25 is $702.6 million — about 2.4 times as much as its pre-9/11 budget. 

 

U.K.’s intelligence partnership with Canada

On 5 December 2023, the Intelligence and Security Committee (ISC) of the U.K. Parliament published a report on the U.K.’s international intelligence partnerships. 

As the report states, the “partnership with the other Five Eyes countries is – and will remain – the most important element of the [U.K.] Intelligence Community’s international engagement.” The U.S. is by far the largest and most important of the Five Eyes partners, but Canada and the other countries also get some attention in the report, and on the topic of Canada the committee makes a couple of intriguing assertions. 

 

More advanced capabilities

One assertion is that some of Canada’s intelligence capabilities are “more advanced” than those of the U.K.:

“The UK–Canadian intelligence partnership is mutually beneficial. Some of Canada’s capabilities – for instance, *** – are “more advanced” than the UK Intelligence Community’s, and CSIS and CSE provide valuable reporting on a range of intelligence topics.” (p. 83)

(The ISC uses three asterisks to indicate where text in the classified version of its report has been redacted.)

Here, the committee was probably thinking mainly of CSE’s Canadian Centre for Cyber Security:

“GCHQ noted in particular Canada’s mature and leading role on cyber security within the Five Eyes: Canada has been with us at the head of the pack on cyber security and our relationship on cyber security is extremely strong and deep. It’s the deepest of the Five Eyes actually and they have pioneered some things that we are using, including how you monitor for threats across government, and similarly we’ve shared capability in the other direction. So I think Canada is really nimble and they’re very focused on cyber security.” (pp. 83-84)

Canada’s development of host-based sensors is perhaps the best known example of Canadian cyber security technology transfer to the U.K.

But the ISC also praises Canadian intelligence collection and/or analysis capabilities. Unsurprisingly, however, on that subject they leave us guessing as to what exactly Canada is good at: “GCHQ also highlighted Canada’s “very good analytical understanding ***” and the fact that, ***, Canada has become a “world leader in ***”.” (p. 84)

 

The sun never sets on the UKUSA empire

Also interesting is the report’s description of a Five Eyes practice known as “follow the sun,” a term that I hadn’t seen before.

The Chief of Defence Intelligence “explained to the Committee that geography enables the Five Eyes to achieve more complete and consistent intelligence coverage. This is done through so-called ‘follow the sun’ working, whereby high priority tasks can be ‘passed’ around the Five Eyes community to allow 24-hour working: imagery analysts that are at [RAF] Wyton … hand the mission on to analysts that are in Washington D.C. or St Louis and to our Canadian partners who will then hand on to Australian and New Zealand partners who then hand back to us” (p. 65)

The Canadian partners specifically referred to here are presumably analysts at the Canadian Forces Joint Imagery Centre, which is part of the Canadian Forces Intelligence Command (CFINTCOM).

The report also describes a new building at RAF Wyton that hosts not only British imagery (and other INT) analysts, but also analysts from the other Five Eyes countries.

“[Defence Intelligence] has taken significant steps to integrate Five Eyes partners into its work through the development of its ‘Pathfinder’ facility at RAF Wyton. This was described to the Committee as “a unique experiment within the Five Eyes community … designed from the outset to accommodate Five Eyes working, both in terms of having Five Eyes personnel on the floorplate ***”.” (p. 63)

Note the flags of all five countries flying at the entrance to the building.

The Pathfinder building was described in 2014, one year after its construction, as “headquarters to [the U.K.’s Joint Forces Intelligence Group] and home to the Defence Intelligence Fusion Centre (DIFC).”

“Within the new Pathfinder Building at Wyton, the Joint Intelligence Operations Centre (JIOC) coordinates Intelligence, Surveillance and Reconnaissance (ISR) for Defence, while the DIFC brings together Geospatial intelligence for the Nation and Intelligence Fusion for Defence.”

The image below is an architect's rendering of the building's interior.

Whether Canada has personnel in the building and, if so, whether it is one or two analysts working on exchange or a somewhat larger formal detachment has not been revealed.

I would assume that the SIGINT part of the Five Eyes partnership also benefits from the use of follow the sun working, although the partner agencies focus their efforts on national priorities and also divide some of their work according to geographic and topical specialties. 

The Five Eyes SIGINT partners also exchange personnel and, in some cases, have deployed detachments to work at partner facilities.

In Canada’s case, CSE has a liaison office and also some personnel serving on exchange with GCHQ, and normally at least two members of the Canadian Forces Information Operations Group are on exchange at RAF Digby, home of the Joint Service Signals Organization, JFIG’s SIGINT component.

With respect to the Canada-U.K. SIGINT relationship more generally, the committee concluded that “it is apparent that GCHQ’s partnership with its CSE counterparts – described as “flourishing” to the Committee – is particularly strong.” (p. 83)

 

Canada joined UKUSA in 1948?

Another interesting claim in the committee’s report is that Canada joined UKUSA in 1948.

“The 1946 British–US Communication Intelligence Agreement – subsequently known as the UKUSA Agreement – formalised the SIGINT partnership and committed the UK and US to an unprecedented level of peacetime co-operation. The UKUSA Agreement was subsequently extended to include Canada (in 1948) and Australia and New Zealand (in 1956), thereby creating the ‘Five Eyes’ intelligence-sharing alliance.” (p. 61)

“Canada was the third country to join the UKUSA intelligence-sharing Agreement (in 1948), and as such is one of the UK’s oldest intelligence partners.” (p. 83)

Similar accounts have cropped up before.

In its 2010-11 report, the Security Intelligence Review Committee summarized the history as follows:  

“During the Second World War, Britain and the United States worked together closely in intercepting the communications of their enemies—what is commonly known as signals intelligence. In 1946, in the context of the emerging Cold War with the Soviet Union, the two major powers decided to institutionalize this cooperation through a formal agreement. Two years later, Canada joined this alliance, with New Zealand and Australia following suit in 1956.” (p. 20)

However, in 1990, Canada’s Intelligence Advisory Committee provided a rather different take on this history in its classified overview of the Canadian intelligence community:

In this telling, the BRUSA (later called UKUSA) agreement was signed in 1945; Canada, Australia, and New Zealand all agreed to join the accord in 1946; and formal agreement incorporating all five parties was reached in 1948.  

As far as I can tell, none of these accounts is quite right.

Let’s start with the IAC.

·        While the text of the BRUSA agreement had been agreed by its U.S. and U.K. negotiators in virtually final form by November 1945, it was formally signed by the two governments on 5 March 1946. Unless there was also an earlier signing, it’s just wrong to say it was signed in 1945.

·        The BRUSA agreement included provision for participation by Canada and other Commonwealth dominions in the partnership if they pledged to abide by certain key clauses of the agreement, and Canada, Australia, and New Zealand provisionally did so at the Commonwealth SIGINT Conference held in February-March 1946. So, if you’re flexible about what it means to “join” the agreement, that part of the IAC account is right.

·        In December 1947, the U.S. and U.K. moved to further formalize the status of the Commonwealth collaborators, limiting the anticipated pool to the existing three countries and laying out the requirements those countries would have to follow in a new appendix to the BRUSA agreement, Appendix J. As part of those requirements, the two parties agreed that the U.K. would “obtain from the Sigint authorities of the collaborating Dominions formal assurance that they will abide by the terms of paragraphs 5, 8, and 9 of the [BRUSA] Agreement and of paragraph 5 of Appendix E to that agreement.” In February 1948, the U.K. conveyed those assurances to the U.S., and when the U.S. and U.K. updated the BRUSA appendices in July 1948 a footnote to that effect was added to Appendix J. These developments are probably the basis for the claim that a formal five-party agreement was signed in 1948. But, in fact, the BRUSA agreement remained a two-party agreement.

There is no evidence, or at least no evidence that I’m aware of, that any actual five-party agreement that could be considered in any way comparable to the BRUSA agreement was signed in 1948 or indeed at any time during the Cold War. The two-party BRUSA agreement was still a two-party agreement at the time it was renamed UKUSA in 1952, and it remained a two-party agreement for the decades that followed. The other three countries were members of the UKUSA partnership, were considered Second Parties, and even had some say over portions of the agreement that affected them, but they were never signatories, and relations among the partner agencies were mostly bilateral rather than Five Eyes-wide.

According to an NSA document leaked by Edward Snowden, “Although NSA has had bilateral relationships with individual Second Party countries going back to the 1940's and 1950's, we did not have any group (5-EYES) partnership until 1993." 

It is only since the 1990s that the Five Eyes partnership has operated on the basis of regular five-party meetings and governance structures.

Another notable piece of negative evidence can be found in the chronology of significant events recorded in the official History of CBNRC (CBNRC was CSE’s name prior to 1975). The page for 1948 contains one redacted entry and a couple of other minor redactions, but there is nothing that might represent a major five-party SIGINT agreement. (The redacted entry is cited to paragraph 9 of chapter 5, which discusses the competition between strategic and tactical priorities in the naval intercept and direction-finding program. SIGINT partners and agreements are mostly discussed in chapter 11.)  

The author of the SIGINT portion of the history, former CSE Chief Kevin O’Neill, was already a senior CBNRC official in 1948, so it is hard to believe he would have failed to include an agreement of such importance in this chronology if a formal quinquepartite agreement actually existed.

Still, for what it’s worth, a CSE document from 2013 makes an almost identical claim, minus the five-dollar word:

So, what’s going on here?

With a vigorous amount of mental contortion, it is possible to interpret these statements not as claims that a single, specific five-party accord was formally agreed in 1948, but as acknowledgments that by 1948 all five countries had in some manner formally agreed to abide by certain key BRUSA provisions. Arguably, that was a development of some significance, although evidently not enough to impress Kevin O’Neill.

Maybe that’s all that those statements were intended to mean. In their plain readings, however, they just seem to be wrong. 

(That said, I’d be happy to see evidence to the contrary.)

Let’s turn now to the U.K. Intelligence and Security Committee’s account.

The ISC, you’ll recall, reported that Canada joined UKUSA in 1948, but that Australia and New Zealand didn’t do so until 1956. Clearly, then, the committee did not have the formal assurances that the three countries provided to the U.K. in 1948 in mind. And the ISC certainly makes no mention of any five-party accord in that year.

The big event that occurred in 1956 was the formal approval by Australia and New Zealand of the conditions for SIGINT cooperation with the U.S. that were negotiated at and in the wake of the Melbourne Tripartite Conference of September 1953 and then added to the UKUSA agreement as Annexure J1 in 1955.

At the time, Australia’s Defence Signals Branch (DSB) was a joint British/Australian/New Zealand organization staffed mainly by Australians but with a significant British contingent and also some New Zealanders. (New Zealand had a radio intercept program and contributed personnel to the DSB, but it had no SIGINT agency of its own until 1977.) The 1956 developments were the culmination of the DSB’s transition from an organization whose only connection to U.S. SIGINT was through GCHQ to one that maintained direct contacts and undertook joint work with the U.S. as well as the U.K.

That’s a significant milestone, and perhaps a reasonable place to identify as the point where Australia and New Zealand truly became UKUSA partners as opposed to just British partners.

But if that’s the criterion we use to judge Australian and New Zealand participation in UKUSA, what does the 1948 date ascribed to Canada correspond to?  

It might be argued that the CANUSA agreement, which formalized SIGINT relations between Canada and the United States, is the logical counterpart to the DSB’s 1956 transition.

But the CANUSA agreement was finalized in 1949, not 1948. So, that doesn’t explain the ISC’s date.

I can’t find anything that does explain it. Just as there is nothing in the History of CBNRC chronology for 1948 that corresponds to a five-party SIGINT agreement, there is nothing that suggests that the UKUSA agreement was “extended to include Canada” during that year.

So, it’s a mystery — and one that also applies to the SIRC account quoted above.

The idea of picking a single date may be part of the problem here.

The integration of Canada, Australia, and New Zealand into the UKUSA partnership is probably best understood as a gradual process of growing interaction and deepening ties that played out over decades — a process that arguably is still playing out as the nature of the partnership itself continues to evolve — so it is quite difficult and perhaps not very helpful to choose a specific date and proclaim it as the line between in and out.

Even what seems like a single event can in fact extend over several years. For example, the History of CBNRC dates Australian (and by implication New Zealand) entry into the partnership to September 1953, the date of the Melbourne Tripartite Conference, arguably the key decision point in the process of integration that culminated in the 1956 formal approval of UKUSA annexure J1 that the ISC report identifies as the moment of partnership.

In Canada’s case, if we do want to pick a year, I think neither 1948 nor 1949 is the one we ought to be looking at.

In my view, Canada was a participant in the post-war UKUSA partnership from the start.

Canada’s residual wartime cryptologic units, consolidated into the Joint Discrimination Unit (JDU) at the beginning of August 1945, remained in operation when the war ended, evolving a year later into the Communications Branch of the National Research Council (CBNRC), which ultimately became CSE. A number of intercept stations also remained in operation, although with reduced staffs. The Canadian government retained these facilities with the explicit expectation that the Canadian post-war SIGINT effort would operate as part of a greater U.S. and U.K. partnership.

Even as they were negotiating that post-war partnership, the U.S. and U.K. also anticipated that combined work with Canada would continue. By November 1945, following consultations with Canadian officials, they had written that expectation into the provisional text of the BRUSA agreement. Canada’s anticipated role was also recognized in the recommendations of the Commonwealth SIGINT Conference held in February-March 1946, which acknowledged that Canada would not be working solely with Commonwealth partners:  

"Canada's geographical position and treaty relations with the U.S.A. make it necessary for her to be able to work directly with Washington. The allocation of interception and cryptographic tasks and the dissemination of Signal Intelligence results to Ottawa will therefore be matters for consultation between Signal Intelligence authorities in Ottawa, Washington and London."

Interim post-war assignments had already been allocated to Ottawa by both London and Washington by that time. While Canada’s intercept and cryptologic efforts were both very small, they were at least marginally operational, worked directly with their U.S. and U.K. counterparts, and were in the process of becoming more capable. In January 1947, CBNRC began working on one of the partnership’s core efforts, processing a portion of the traffic carried by one of the Soviet Union’s most important high-level cryptographic systems, the teleprinter the UKUSA partners called Coleridge. The annual report of the U.S. Army Security Agency (ASA) for fiscal year 1947 (i.e., July 1946 to June 1947) noted that “channels of liaison with other communications-intelligence agencies were gradually systematized and standardized by implementation of joint processing agreements with the British (LSIC), the Navy (CSAW), and the Canadians (CBO)."

As time went on, Canada became increasingly integrated into the partnership, exchanging liaison officers with the U.S. and securing improved access to U.S. SIGINT reporting after the 1949 CANUSA agreement, for example. The nature of Canada’s participation in the partnership has extended and deepened in many ways in the decades since. 

But the fundamental partnership decision had been made by all three parties by the fall of 1945: although never a signatory of the BRUSA/UKUSA agreement itself and always a junior player, Canada was a partner from the beginning.

Redactile dysfunction

Anyone who uses Canada’s access to information process on a regular basis will likely come to suspect that information is occasionally withheld not for national security or other legitimate reasons, but to protect the institution withholding the information from potential embarrassment or controversy. This is not a legitimate reason for withholding information under the access law, but since the information in question has been, well, withheld, it is very difficult to detect this unlawful practice when it may occur. Normally.

Occasionally, however, we do get to see the information that lies behind the redactions — and, perhaps unsurprisingly, in some of those cases our suspicions do seem justified.

Case in point: In 2021, NSIRA released the declassified version of its Review of CSE’s Self-Identified Privacy Incidents and Procedural Errors (NSIRA Review 08-501-2). This was not an access to information release as such, but the report had been subjected to a comparable redaction process overseen by CSE prior to its release.

Among the items redacted from NSIRA’s report were the details of one of the practices CSE had been using to mitigate privacy incidents during which CSE or one of its Second Party partners had inadvertently identified a Canadian person or institution in SIGINT reports.

NSIRA did not believe that the practice was an appropriate way to respond to such cases and recommended that it be rescinded.

I was reminded of this report recently when I was going through CSE’s response to access to information request A-2022-00029, in which the original requester had sought the release of a series of briefing notes written by CSE for the Minister of National Defence.

One of those briefing notes concerned NSIRA’s review and the steps that CSE was taking to respond to it. It too redacted all details of that particular mitigation technique, even though CSE had abolished the practice in response to NSIRA’s objections, and had in fact done so even before NSIRA had finalized its report.

What was this inappropriate practice which, even after its abandonment, was too sensitive for NSIRA to be permitted to disclose?

The NSIRA report did manage to retain one clue. A flowchart of CSE’s process for responding to privacy incidents reproduced at the end of the report contained this intriguing, partly redacted step possibly relevant to NSIRA’s concerns:

To get the full answer, however, you have to turn to the documents that were provided to the British Columbia Civil Liberties Association (BCCLA) during its recent litigation against CSE. (Those documents were also subjected to a stringent redaction process, but — possibly because they were provided for use in the Federal Court — the redactions were often less sweeping than those made to documents released through the access to information process.)

According to sections 8.3 and 8.4 of OPS-1-7: Operational Procedures for Naming in SIGINT Reports (provided to the BCCLA as document AGC 0019),

Generally speaking, you must cancel and reissue reports in which you inadvertently named or contextually identified Canadians or Second Parties. However, after issuing a number of reports (more than 10) in which you named or contextually identified Canadians or Second Parties you believed to be foreign, you may learn that the person, corporation or organization is actually Canadian (or Second Party). In this case, you may be able to obtain retroactive blanket approval for these historical reports.

You must contact Operational Policy, who will assess retroactive blanket approval requests on a case-by-case basis. The Director, COP is responsible for granting retroactive blanket approvals. With a retroactive blanket approval you do not need to cancel or reissue these 10 or more historical reports since this might draw unwanted attention to the inadvertently identified Canadian.

 

It was this policy of retroactive naming approval that NSIRA objected to, and that CSE agreed to rescind but evidently wanted to keep secret.

 

Why keep this information secret?

Nothing about this information reveals even the slightest clue about what CSE and its partners seek to collect intelligence on, how they go about collecting intelligence, how their own security might be compromised, or anything else of a legitimately sensitive nature.

All it does is tell Canadians that if CSE or its partners name or contextually identify you in 10 or more reports under the mistaken belief that you are not Canadian, and then later they learn the truth, the response in at least some cases in the past was to retroactively approve the mentions in the old reports on the rather doubtful grounds that dredging up those reports to either cancel or correct them might draw more attention to the individuals inappropriately named than leaving them untouched in the files.

Is it embarrassing for CSE that it used to embrace a policy that both NSIRA and now CSE itself recognize was not an appropriate way to protect Canadians’ privacy? Potentially, sure, yes.

Is it embarrassing that CSE needs to have a policy for what to do when it or its partners mistakenly identify a Canadian on 10 or more separate occasions? Yeah, that too. There are reasons why that sort of error happens, but it’s certainly not a great look.

This is not the only case in which CSE redacted information for no evident reason other than it was potentially embarrassing.

The agency also did it to this recommendation from a June 2006 report by NSIRA’s predecessor, the CSE Commissioner:

Once again, we know the missing words thanks to a document provided to the BCCLA (AGC 0260):

The text that CSE redacted was: “and ensure that all decisions and resulting activities are based upon criteria that have been consistently applied and are statutorily defensible.”

Is it embarrassing for CSE that the CSE Commissioner felt obliged to recommend that all the agency’s decisions and actions be based on consistently applied and statutorily defensible criteria? Arguably, yes, since it implies that CSE had sometimes failed to do so in the past.

But is there any way in which redacting those words from a document obtained through the Access to Information Act could be construed as itself statutorily defensible? Not a [redacted] chance.

The Access to Information Act is a law, and it applies to CSE as much as to any other federal government institution. That act gives those institutions an enormous amount of discretion in deciding what information to release and what to withhold, but it does not give them the option to withhold information just because they think it could make them look bad.

Let me repeat that. Potential embarrassment is NOT A LAWFUL JUSTIFICATION FOR REDACTING INFORMATION requested under the Access to Information Act. It’s just not. 

It’s not.

How often does this sort of redaction get made? Who knows?

Is it too much to ask that CSE hold itself — or, if necessary, that NSIRA hold CSE — to the standard of compliance with this law?

 

RIP David Kahn

U.S. journalist and author David Kahn died on January 24th. He was 93.

Kahn's 1967 book The Codebreakers, which covered the history of code making and breaking from its ancient origins to the mid-20th century, thrust the previously little known field of cryptology into the public eye to an unprecedented degree and thus helped open the era of public discussion of signals intelligence.

Preferring its comfortable obscurity, the U.S. National Security Agency feared the impending publication of the book so much that it looked for ways to prevent its publication. Fortunately, it was unsuccessful. In the years since then the agency has gradually come to recognize the value of greater public awareness of its work. In 2020 it inducted Kahn into its Cryptologic Hall of Honor. 

In addition to writing The Codebreakers, Kahn was one of the founding editors of the journal Cryptologia. In 1991 I sent him a draft of an article that I had written about CSE's 1980s cryptanalytic renaissance, and when he very graciously replied it was with a copy-edited version accepted for publication in the journal. "The Fall and Rise of Cryptanalysis in Canada" appeared in the January 1992 issue. You can read a slightly updated version I posted on this blog here.

The New York Times published an excellent obituary of David Kahn on February 9th.

R.I.P.

One small step for transparency...

...one giant leap for common sense? 

Possibly.

SLINGSHOT is CSE's "SIGINT production and dissemination system. It is used for gathering client requirements; end-product report (EPR) authoring, storage and searching/retrieval (including Second Party reports). It also allows for monitoring and logging of client access to EPRs."

For two decades, CSE has insisted on redacting SLINGSHOT's name whenever it appears in documents released to the public — as in the example shown above — despite the fact the name has been publicly known for almost all those years. 

It looks like that policy may finally have changed.

SOMOS for that secret

As long ago as 2006, the consulting firm that helped develop the system revealed its name in a web post:

The system was also mentioned in a number of the Five Eyes documents that Edward Snowden leaked in 2013. For example:

Those revelations were of course unauthorized, but the name was also confirmed in a number of official document releases when the redactors evidently failed to catch it.

Such as:

And: 

Still, the blanket redaction policy went on. 

This example, and it's just one of many, is from an NSIRA report released in 2021:

 

Schrödinger's CATAPULT

That report, incidentally, contained my favourite redaction of all time, a sort of quantum superposition in which the word SLINGSHOT was redacted for anyone who didn't know what it was but was obvious to anyone who did:

It's both there and not there at the same time!

SLINGSHOT comes into the light

Has the madness finally ended? Maybe.

NSIRA's report on CSE internal sharing of information related to Canadians, released on January 25th, contains not one, but two, unredacted instances of SLINGSHOT's name, suggesting that we may be witnessing a deliberate decision by CSE to declassify the name rather than a routine redaction fail.  

 

If this really is a change in policy, I'd like to think my little rant on Elon's Hell Site back in November had something to do with it. Maybe it gave the folks at NSIRA enough ammunition to convince CSE it really did make no sense to continue redacting SLINGSHOT.

If so, it's one small step forward for transparency. 

There is of course a whole lot more about CSE that is well known — or should be well known — that the agency continues to refuse to acknowledge or provide official information about on spurious security grounds. 

Still, every journey starts with a single step.

Update 3 February 2024: 

Yet another instance of SLINGSHOT unredacted turned up just one day after this post was written, in a Federal Court ruling issued on October 10th, 2023, but only made public in declassified form on January 30th.

It seems pretty clear now that this really is a change in policy. Score one for common sense!

The satellite monitoring site that never was

 

 

Was Alberta once considered for the location of a satellite monitoring site for CSE? That’s my current working hypothesis. 

An access to information request I recently submitted to the Privy Council Office may eventually provide the evidence to confirm or reject that hypothesis — but only if Ottawa can transcend its reflex for pointless redactions.

 

During the 1980s, CSE undertook a major effort to modernize the Canadian SIGINT program. Among other initiatives, the agency revitalized its cryptanalytic capabilities, established intercept sites in Canadian diplomatic facilities, began monitoring commercial satellite (COMSAT) communications, and bolstered CSE’s staff by 50%. (You can read more about CSE’s 1980s renaissance here.)

Satellite communications were an increasingly important part of both government and non-government international telecommunications during the 1980s. 

Commercial communications satellite services began in 1965, when an intergovernmental consortium called INTELSAT launched Early Bird, the first commercial communications satellite. Shortly thereafter, NSA and GCHQ set up the ECHELON program to monitor traffic of interest on INTELSAT’s satellites. 

By the 1980s, the growing volume of communications carried by INTELSAT and other commercial and national satellite operators made it desirable to bring the other UKUSA partners into the program. In March 1987, Australia announced plans to construct a satellite monitoring station at Geraldton, Western Australia, and in December 1987, New Zealand announced that it would build a similar station at Waihopai.

For Canada, entry into the satellite monitoring program was understood as a means both of augmenting our contribution to the UKUSA partnership and of collecting intelligence of specific interest to the Canadian government.

Documents recently released to the Canadian Foreign Intelligence History Project (CFIHP) through the Access to Information Act confirm the broad outlines of the Canadian plan. These documents show that the satellite monitoring project was a key element of the renewal plan that CSE pitched to the Interdepartmental Committee on Security and Intelligence (ICSI) in March 1984 in its Strategic Overview of the Cryptologic Program, 1985-1988.

The Strategic Overview document itself is rather heavily redacted, but it does confirm that one of the projects CSE proposed is related to COMSAT collection, and a handwritten annotation notes that this project was approved.

Another document, an External Affairs memo from December 1987, is more revealing, confirming that “ECHELON is a CSE project which was designed to collect Intelsat communications…. Our position on ECHELON has been to support the project as a valuable contribution to the overall Canadian and allied effort.” At the time of that memo, the project was on hold due to legality concerns expressed by the Department of Justice. But those concerns appear to have been resolved not long afterwards, as 1988 documents confirm that the project was back on track. A June 1988 document notes, for example, that “PILGRIM and ECHELON are going forward.” (PILGRIM was the project to operate intercept sites in Canadian diplomatic facilities.) Another document, from March 1988, lists “possible options to address identified intelligence deficiencies," one of which is "greater exploitation of the ECHELON program to yield more Canada-specific information, while contributing to the allied SIGINT effort."

Canadian Forces Station Leitrim, located just south of Ottawa, became the home of Canada's satellite monitoring effort.

Air photos that the author has examined at the National Air Photo Library show that the first satellite monitoring dish was installed at Leitrim between late 1984 and early 1985. A second large dish was installed in 1985-86, followed by a third in 1987 and a fourth in 1989-90. A couple of small dishes were also in place by that time.

 

This 1988 photo, taken from Leitrim Road, shows the three main dishes then at Leitrim (two of them covered by radomes). A small dish can also be seen between the left-hand radome and the large uncovered dish.

 

Another site was proposed

The Strategic Overview document reveals, however, that Leitrim was not originally intended to be Canada’s primary satellite monitoring site. The fact that one or more new facilities were envisaged was redacted from the version released to the CFIHP, but a less redacted portion of the document tells the tale:

The risks associated with this initiative relate to the [redacted.] If this does not happen, a COMSAT training and R&D facility to be developed as part of the project at Leitrim will be upgraded to that of a primary facility. The satellite communications which can be collected from this site represent similarly [redacted.]

Another document released to CFIHP also confirms that “Should [redacted element of the plan] fail to happen, a training site planned for C.F.S. Leitrim will be developed into a full-fledged collection station”.

Where might CSE have wanted to monitor satellite communications originally?

Leitrim is in a good location to monitor the INTELSAT satellites stationed over the Atlantic Ocean, which carry communications between the Americas and Europe/Africa. It could also monitor many of the national satellites that serve parts of the Americas, such as Mexico’s Morelos satellites and Brazil’s Brazilsats, both of which systems were established in the 1980s. But it is too far east to monitor the INTELSAT satellites over the mid-Pacific.

Thus, CSE may have wanted to build a separate West Coast site from which to collect satellite traffic between Asia and North/South America. Or it may have sought a single site from which satellites over both the Atlantic and the Pacific — and everywhere in between — could be monitored. 

 

Alberta bound?

Such a site would have been possible in southern Alberta, although the furthest east of the satellites over the Atlantic and the furthest west of the satellites over the Pacific would not be visible. (The arc of coverage would range from about 175-180 degrees east to 40-55 degrees west.)

Was such a site under consideration? Another document released to the CFIHP suggests it may have been.

The document is a list of intelligence-related files held by the Privy Council Office. Among other topics, the list contains several pages of CSE-related files, including two sets of files, both established in 1986, called “Collection Sites — Alberta”.

It is very unlikely that these files refer to radio collection sites. Canada hasn’t had a radio collection site in Alberta since the Canadian Army’s Grande Prairie site was closed in 1947, and I can’t imagine any reason why CSE would have considered opening a new radio collection site in the province in the 1980s. One of the main goals of CSE’s modernization project was to move the agency away from its overreliance on radio collection: the same year these files were opened the major radio collection site at Inuvik was closed.

Consideration of possible locations for a satellite monitoring site thus seems like a much more likely explanation for these files. In September I submitted an access to information request asking for the records in the files to be released. Now we wait to see what PCO and CSE will agree to release.

 

But why wait?

In the meantime, it’s fun to speculate as to where an Alberta satellite collection site might have been built had the plan gone ahead.

My wild ass guess is that Canadian Forces Base Suffield, the largest army training area in Canada, was CSE’s main candidate. Located about 50 km northwest of Medicine Hat, the 2,700-sq-km base also hosts DRDC Suffield (formerly called Defence Research Establishment Suffield).

Building the site at Suffield would have made the CSE station quite similar to NSA’s Yakima Research Station, one of the first ECHELON sites, which was located at the U.S. Army’s 1,300-sq-km Yakima Training Center in Washington state from 1974 until roughly 2013, when its functions were transferred to Buckley Air Force Base (now Buckley Space Force Base).

CSE may have hoped that if it built the site on a base like Suffield, its true purpose would go unnoticed. Like Yakima, Suffield would have provided a location big enough to keep the dishes largely away from prying eyes on land already owned by the Department of National Defence and with support services already available. Construction of the dishes could have been explained as communications research work associated with Defence Research Establishment Suffield, while the existing civilian and military workforce at the base would have enabled the mostly military intercept staff to hide in plain sight, at least potentially drawing much less attention than a newly constructed free-standing site would have.

The base was also well served by high-capacity telecommunications, being directly on the route of the Trans-Canada Microwave System.

So Suffield seems like a natural candidate.

 

As I said, however, this is all wild ass speculation. It may well be that Suffield was never under consideration. It could even be that the “Collection Sites — Alberta” files are unrelated to CSE’s satellite monitoring proposals.

As far as we know, no collection site of any kind has been built in Alberta since the 1940s. There is some reason to believe that in 1992 CSE investigated the possibility of building a separate satellite monitoring station in Ontario, at the former National Research Council radio observatory site at Lake Traverse, Algonquin Park. But nothing came of that either.

In the end, Leitrim became CSE’s primary satellite monitoring site, and it remains the primary site today. Documents confirm that INTELSAT monitoring associated with the ECHELON program went ahead, but it seems that it did so without the construction of a separate satellite monitoring site in Alberta or anywhere else.

Will PCO and CSE release any additional information that sheds light on what CSE proposed, what did and didn’t occur, and why these decisions were made some 35-40 years ago? That remains to be seen.

 

CSE budget authority tops $1 billion

Planned additions to CSE's fiscal year 2023-24 budget authority that were announced in the Supplementary Estimates (B), tabled in parliament on November 9th, will push the amount of money the agency is authorized to spend this year above the $1 billion mark for the first time in the agency's history.

The changes proposed would result in a $15,196,568 net increase in CSE's 2023-24 budget authority, boosting the total figure from $984,855,602 to $1,000,052,170. 

Notable increases include a $10,771,964 top-up for the ongoing operations of the Canadian Centre for Cyber Security; $1,592,171 for the Interim Quantum Safe Capability project; $1,500,000 for advertising programs; and $1,176,929 to "enhance national security through an academic research initiative" (part of a project originally announced in Budget 2022).

The full list of changes, including an increase in funding for statutory programs and various transfers between departments, can be found in the estimates document. 

Normally, there is a small shortfall between the amount of money CSE is authorized to spend in a fiscal year and the final amount actually spent during that year, so it is possible that the agency's spending ultimately will fall somewhat short of the $1 billion milestone this year. However, there will also be another opportunity for CSE to receive a boost in spending authority before the end of the fiscal year (the Supplementary Estimates (C), expected in February 2024), so the final figure is very much up in the air.

Cyber operations spending revealed

Meanwhile, there has been a change in the way CSE's past spending is reported in the online GC InfoBase. 

In previous years, CSE's spending was broken down into two major programs: Foreign Signals Intelligence and Cyber Security. But the spending for the most recent year for which numbers are available (FY 2022-23) is broken into four programs (click image for a better view): $336,912,405.10 for Foreign Signals Intelligence, $9,145,757.10 for Foreign Cyber Operations, $280,703,287.42 for Operations Enablement, and $304,486,444.45 for Cyber Security.

No explanation of these categories is provided, so we are on our own to interpret what they mean. I think what's going on is this: The Cyber Security program covers the spending of the Canadian Centre for Cyber Security (the Cyber Centre), as it did in the past. The other three programs cover the spending that used to be reported simply as the Foreign Signals Intelligence (SIGINT) program.

The change was probably made to enable spending on CSE's Foreign Cyber Operations, which comprise the Active Cyber Operations (ACO) and Defensive Cyber Operations (DCO) that were added to CSE's mandate in 2019, to be reported as a separate program. The tricky part is that these activities mostly use the same IT systems and knowledge base and even a lot of the same personnel as CSE's SIGINT activities. They also both benefit from the same administrative and security services, maintenance activities, and office accommodations provided by the agency. It looks to me like these common services are now reported as the Operations Enablement program. It's possible that certain common services also used by the Cyber Centre are listed in that program as well, just to make things more confusing, but I'm guessing that probably isn't the case.

The remaining two programs probably list just the resources dedicated specifically to SIGINT and to cyber operations, possibly just the direct personnel costs for the staff assigned to the SIGINT production chain on the one hand and those assigned to ACO/DCO activities on the other, as measured in full-time equivalents (FTEs). 

Whatever their exact composition, the two spending numbers associated with these programs, $336.9 million and $9.1 million, respectively, would appear to indicate that as of 2022-23 the Foreign Cyber Operations program was only about 1/37th the size of the SIGINT program. In FTE terms, that might translate to something like 60 people in cyber operations. (If more than just direct personnel costs are counted for these programs, the number of people involved would be lower, possibly as low as 30 for cyber operations.)

I would offer kudos to CSE for its willingness to see this information published, but frankly I'm going to wait until next year when we see whether this exercise in transparency continues.