Friday, January 27, 2012

OSINT and CBNRC

As the Gouzenko revelations, the allegations in the headlines of today, and innumerable incidents in between demonstrate, the Russians love to spy. But does that love of the excitement, intrigue, and duplicity of traditional spycraft blind them to the value of open source intelligence? And did it blind them in the past?

Few Canadians had heard of the Communications Branch of the National Research Council in the late 1940s and 1950s, and even fewer knew anything about it. But the Soviet intelligence services knew of it, maybe through Canadian sources and certainly through their agents in the U.S. and the U.K.

And with nothing more than knowledge of its name, the Soviets could have learned a great deal about CBNRC through open sources – assuming they were smart enough to check them out.

Let’s see what we (and they) could have learned using just the information available to someone in the public library in 1955.

NRC Review

The first place we’ll check is the NRC’s annual public report, the 1955 edition of the National Research Council Review. Ah, this is promising: there’s a list (pp. 247-267) of the top 550 or so NRC employees by name, NRC division or section, and job class.

Sadly, however, the Communications Branch doesn’t appear anywhere in the annual report. It’s treated as if it doesn’t exist at all.

So we move on.

Public Accounts of Canada

Next we check the 1954-55 Public Accounts of Canada, the annual official record of federal government spending.

Now, this is interesting: there’s a list (pp. P-10 to P-13) of the 380 or so NRC employees who earn $5000 or more a year.

Maybe the NRC Review will come in handy after all. Let’s compare the lists.

It takes quite a while to cross all of the names in the NRC Review off the list in the Public Accounts (believe me, I’ve done it), but, hmmm, the results are quite interesting: roughly 35 of NRC’s top-earning employees don’t appear anywhere in the NRC annual report.

Some of this discrepancy can be put down to new arrivals who don’t make it onto the annual report list, some to departures during the year, and some perhaps to clerical errors. But a check of earlier reports shows that most of the names, some 25 of them, never appear in the annual reports.

So who are these people? Let’s have a look at their names:
Browness G A
Carson W S
Chramtchenko M
Colls T G S
Denning C E
Diditch S J
Dornan J E
Drake E M
Ensell G
Featonby J
Handforth R E
Hellyer C N
Hepburn S K
Johnston P A E
Jury J M H
MacAskill R
MacKiddie C G
Maillet R J
McLaren R S
Odin J P
Oliver M S R
O'Neill N K
Thomson G S
Trowbridge W J
Wilkins T J
The Public Accounts list provides the annual salaries of these people. Salary is normally a pretty good indication of rank, so let’s sort the names by income:
Drake E M [$10.5 k]
O'Neill N K [$8.2 k]
MacAskill R [$7 k]
Diditch S J [$6.56 k]
Colls T G S [$6.4 k]
Hellyer C N [$6.36 k]
Trowbridge W J [$6.32 k]
Denning C E [$6.08 k]
Dornan J E [$6.08 k]
McLaren R S [$5.97 k]
Oliver M S R [$5.97 k]
Ensell G [$5.75 k]
Browness G A [$5.72 k]
Featonby J [$5.55 k]
MacKiddie C G [$5.55 k]
Maillet R J [$5.35 k]
Thomson G S [$5.35 k]
Wilkins T J [$5.35 k]
Hepburn S K [$5.33 k]
Chramtchenko M [$5.23 k]
Odin J P [$5.23 k]
Carson W S [$5.15 k]
Handforth R E [$5.15 k]
Johnston P A E [$5.15 k]
Jury J M H [$5.11 k]
By this point my more loyal readers (assuming I have any) will probably have recognized some of these names. There may be one or two mistaken identifications, but aside from those few cases they are all CBNRC employees. Here is the first part of the second list again, with a few annotations added:
Drake E M [$10.5 k] = Ed Drake, CBNRC Director
O'Neill N K [$8.2 k] = Kevin O’Neill, Co-ordinator Production
MacAskill R [$7 k] = Rod MacAskill, Co-ordinator Administration
Diditch S J [$6.56 k] = Steve Diditch, Head Reporting
Colls T G S [$6.4 k] = Tom Colls, Head O Group
Hellyer C N [$6.36 k] = Chuck Hellyer, Head IBM Group
Trowbridge W J [$6.32 k] = Bill Trowbridge, Head T&D Group
Denning C E [$6.08 k] = Cecil Ernest Denning, Head R&D Group
Dornan J E [$6.08 k] = Jack Dornan, Head P Group
McLaren R S [$5.97 k] = Robert McLaren, Head ? Group, former liaison officer to AFSA/NSA
Oliver M S R [$5.97 k] = Mary Oliver, Head Administrative Services
By subtracting one list from the other we have managed to recover the entire top echelon of the CBNRC! (The additional information about their specific jobs, of course, is not revealed.)

We have also identified roughly 6% of CBNRC’s entire staff at that time (albeit with a few possible mistakes), and by examining the distribution of salaries we can even develop a crude sense of the agency’s upper rank structure. Repeating the process using earlier editions of the NRC Review and the Public Accounts going back as far as 1946, we can increase the list of known staff members to more than 50, accounting for more than half of the agency’s staff during its earliest years (and, of course, all of its more senior staff).

Peter Dwyer

One of the names appearing on the earlier lists is P. Dwyer. Although unknown to the average Canadian at the time, this is a name very well known to Moscow: Peter Dwyer was the MI6 liaison officer in Washington prior to Kim Philby’s appointment to the job in late 1949. Philby knew Dwyer well, and he knew that when Dwyer left Washington it was to take a job with the Canadian government in Ottawa. He may even have known that Dwyer’s new job was to be Head of Reporting at CBNRC. Dwyer stayed at CBNRC for only two years, moving from there to the Privy Council Office and eventually to the Canada Council. But for those two years his name appears on our OSINT lists.

Thus, if the Soviets are doing OSINT at this time, Dwyer's name (and possibly others they might recognize through intelligence sources) confirms for them that they are indeed looking at a list of considerable interest.

The Estimates

A look at the government's spending estimates also reveals some interesting information. The 1955-56 Estimates show that NRC had 2198 "Full Time Positions" as of 31 March 1955, whereas the 1954-55 Estimates had shown 2618 "Full Time Positions" as of the same date. What happened to 420 NRC employees? Comparison of the numbers shows that the missing employees have been retroactively transferred (for accounting purposes) to a catch-all category called “Casuals and Others”.

The figures provided in the Estimates enable us to further break down the 420 positions by job category:
Scientific & Executive
1 Director
2 Senior Research Officers
6 Associate Research Officers
80 Assistant Research Officers
38 Junior Research Officers
[Sub-total: 127]

Service Staff
2 Principal Clerks
3 Clerks, Grade 4
26 Clerks, Grade 3
51 Clerks, Grade 2B
40 Clerks, Grade 2A
[Sub-total: 122]

Technical Staff
1 Technical Officer, Grade 3
9 Technical Officers, Grade 2
43 Technical Officers, Grade 1
82 Senior Laboratory Assistants
36 Laboratory Assistants
[Sub-total: 171]

TOTAL: 420
It is immediately obvious that this is no random group of summer interns or other casual employees; it is clearly a coherent organization with its own internal structure. It is CBNRC. The entire organization.

We can speculate as to why the accounting change was made. CBNRC was undergoing substantial growth during the 1950s, and this had the effect of making NRC’s overall staff numbers grow in a way that almost certainly became increasingly difficult for NRC officials to explain and justify to members of parliament and the public. The change was probably intended to obscure the extent of this growth and thus deflect embarrassing questions.

That it also reveals a detailed snapshot of the CBNRC’s staffing and organization as of 31 March 1955 is presumably an unintended side effect.

Liaison officers

That's not all we can find. The same documents that record employee salaries also record the amounts of travel and representation allowances paid to employees. By comparing this data to the public lists of Canadian diplomats posted abroad, it is possible to determine that CBNRC began sending representatives to London and Washington earlier in the 1950s. In the mid-1950s, CBNRC’s representative at GCHQ is J.A. [Joe] Gibson and its representative at NSA is H.M. [Howie] Harris.

Department of Public Works annual report

There is also some information available about the location of the CBNRC’s offices. In an extremely rare public mention of the agency’s name, the 1950 annual report of the Department of Public Works reports that "extensive alterations and improvements" have been done to the Rideau Military Hospital "to provide accommodation for the Communications Branch and Radio-Electrical Engineering Division" of NRC. A quick check with the Radio Electrical Engineering Division, which makes no secret about its location, confirms that it remains at the NRC’s Montreal Road campus, but CBNRC is another matter. No one is willing to comment on that. However, a simple drive past the compound would confirm that the former military hospital (and former convent before that) is now a high-security site, with a guardhouse to control entry at the gate and a fence topped with barbed wire surrounding the grounds. Our OSINT analyst would also be able to see that the size of the building is about right for an organization of some 400-500 people.

OSINT overview

To sum up, during the 1950s, an OSINT analyst starting with nothing more than the name of the CBNRC and a general sense of what to look for could have determined the size of CBNRC’s staff in 1955, the location of its office building, and the names and salaries of more than 50 of its employees, including its entire top echelon. The analyst could also have developed a crude picture of the organization’s structure and determined both the fact that CBNRC had liaison officers in the U.S. and the U.K. and the names of those liaison officers.

In addition, by watching the progression of individual salaries over time it would have been possible to identify whose careers were on the fast track, whose careers were stagnating, and who had just left the agency for some reason.

Such information would have been of tremendous use to the Soviets at the time, both in assessing CBNRC’s capabilities and in identifying CBNRC employees to target for surveillance or potential recruitment. Occasional surveillance of the building would have enabled them to keep track of the size of CBNRC’s staff and also to identify many of its lower-paid employees, who would not have appeared in the Public Accounts lists.

OSINT concerns

The Canadian government was concerned about the dangers of open-source information during the 1950s.

According to Mark Kristmanson,
In October 1952, the Globe and Mail disclosed that for five dollars any foreign intelligence agency could obtain the Canada Air Pilot and discover the location of every Canadian airfield, including the secret ones in the North.... [Peter] Dwyer [who was by then with the Privy Council] soon received a report from the Department of Mines and Technical Surveys that a suspicious-looking man with a Russian accent had attempted to purchase the said publication. The man was Mr. Ogorodnikov, the Ottawa representative of Tass, the official Soviet news agency. Half a century later the incident seems comical, but it prompted serious discussions about restricting public access to open-source information. Air Intelligence Chief Edwards thought it ‘one of those problems peculiar to our democratic way of life’. George Glazebrook advocated a principle of step-to-step reciprocation with the Soviets. RCMP counter-intelligence officer Terry Guernsey allowed that other Russian attempts to purchase the Air Pilot had been monitored and he suggested the whole question of open sources be put up to the Security Panel. Reading these responses, Dwyer... recommended better measures to enhance feedback when Soviet agents attempted to acquire public documents. (Mark Kristmanson, Plateaus of Freedom: Nationality, Culture and State Security in Canada, 1940-1960, University of Toronto Press, 2003, p. 119)
Public documents such as the Public Accounts, the Estimates, and departmental annual reports could not be withheld from the public, however. They could be found in any major public library. The only protection against the loss of information through documents such as those would have been to prevent it from appearing in them in the first place.

It is not clear whether the government ever realized the extent of the information that it was leaking about CBNRC. Some of the information provided was deliberately obscure. Some information, such as most budget data, was successfully hidden. But there is no sign that the government ever made a serious attempt to prevent the kind of information described in this post from appearing. The Public Accounts continued to list public servant salaries until 1964, when, ironically, privacy concerns rather than security concerns finally brought an end to the practice. (It is perhaps worth remembering that lower privacy doesn't always mean greater security.)

Another question remains: Did the Soviets ever conduct OSINT collection on CBNRC? Did they ever take advantage of this excellent source of information? Perhaps they had such good clandestine sources of information about the organization that they had no need to waste time compiling open-source information. There is no public information that suggests they ever had a spy within CBNRC, however. Perhaps they were so in thrall to the supposed romance of espionage that they never bothered to pursue the open sources available.

Or maybe they collected it all, and more. We just don't know.

Unfortunately, this question is one that open sources can't answer.

0 Comments:

Post a Comment

Links to this post:

Create a Link

<< Home