I wrote this a while back, too.

I’m finally getting around to posting a tidbit I found a while back in Tyler Drumheller’s On the Brink: An Insider’s Account of How the White House Compromised American Intelligence. 

I am referring to the assertion that, when Drumheller was a CIA operative in South Africa, his sources

provided incontrovertible evidence that the apartheid government had in fact tested a nuclear bomb in the south Atlantic in 1979, and that they had developed a delivery system with assistance from the Israelis. 

I really wish Drumheller had provided more detail about this evidence, which sounds a lot more concrete than the Vela satellite evidence. Obviously, a debate still rages about South Africa’s alleged test and cooperation with the Israelis, so more information would be welcome.

I thought I remembered writing this:

In Tyler Drumheller’s On the Brink: An Insider’s Account of How the White House Compromised American Intelligence, the author (as many of you know) recounts the contentious email discussion within the IC about the validity of the Iraqi informant named Curveball. 

The whole tale is pretty entertaining, but I particularly like the part where he mentions that he once had to tell one of his colleagues “that it might not be such a good idea to put ‘fuck you’ in an email.”

Happy Monday.

The appendix to this 2004 NPR article contains a South African document which cites the possibility that a terrorist organization “could acquire and launch against us a tactical nuclear weapon” acquired from China.

The authors explain here:

<snip>

<snip>

The article makes clear that this view appears not to have been widespread within the South African government:

I ran across this post based on a series of 2010 interviews with Tyler Drumheller in which he discussed South Africa’s nuclear weapons program:

During our interviews, he reflected on earlier experiences in his career. In 1980, “we had 11 case officers in South Africa, four of them with deep cover.” They had penetrated the apartheid regime’s “Project Circle,” which was already within reach of perfecting a usable, deliverable atomic bomb.

“We were regularly able to obtain swipe samples from its enrichment facilities. We could monitor progress,” Drumheller told me.

They also monitored the deep cover “black station” of Israel’s intelligence agency, the Mossad. Drumheller said this was a front company called TamCo. The Israelis had no idea they were under such close U.S. scrutiny.

Washington had tried to impede Pretoria’s effort by embargoing the shipment of a VAX computer from Massachusetts-based Digital Equipment Corporation. “Project Circle needed that VAX” to complete the project, Drumheller said.

The CIA station in Pretoria learned that South Africa was able to get around the embargo by having the computer — the same powerful VAX model — transshipped from the United States. “It came via the Israelis and TamCo,” Drumheller said.

[snip]

The CIA was able to obtain these and other details about the Israeli-South African collusion, Drumheller explained, from a stellar agent that the CIA had placed within Project Circle.

Here’s more from the PBS interview with Paul Pillar that I cited the other day. He discusses the politicization of intelligence.

A lot of intelligence analysts were caught up in several things: a previous consensus against which there just wasn’t enough intelligence to challenge it; the consensus being that yes, there were programs. The atmosphere in which they were working, in which a policy decision clearly had already been made, in which intelligence was being looked to to support that decision rather to inform decisions yet to be made, was a very important part of the atmosphere.

Exactly how that may have affected the individual judgments of particular analysts, it’s impossible to say. It probably had some effect, particularly since most of the shortcomings of the analysis we’re talking about come down to matters of nuance, caveat — whether the language is too strong, that sort of thing. There were many, many opportunities for things to be shaded in the preferred direction rather than in another direction.

<snip>

Politicization, real politicization, rarely works that way; that is to say a blatant, crude arm twisting. It’s always far more subtle. It would take the form either of these almost subconscious or subliminal adjustments that dozens of analysts might make in the course of phrasing their judgments, making it a little less nuanced, a little less caveated, which I think is the main basis for criticizing the judgments on weapons of mass destruction in Iraq.

It can take the form of … intelligence assessments that conform with what is known to be the policy having an easier time making it through … than assessments that don’t supply the policy. … This wasn’t an inquiry into how can Iraq threaten the United States; it wasn’t an inquiry into what are Al Qaeda sources of support. It instead was basically research in support of a specific line of argument. That, I think, qualifies for the label “politicization,” even if analysts are doing their best job to maintain their analytic integrity when they make their individual judgments. …

This description echoes other assessments of the politicization of intelligence that most of us have likely read. But I think it’s good to have another data point, especially from Dr. Pillar.

This is a CIA “assessment of what occurred at the Novaya Zemlya test site area” in August 1997. Here’s Tenet’s summary:

But I think this part is pretty interesting…if I’m reading it correctly, the below excerpt explains agency procedures which, at least at the time, required analysts to err on the side of caution when assessing possible nuclear tests.