not all. But I think he misses the bigger picture. CIA, since it was created, has been built around 3 major directorates - Intelligence (DI), Operations (DO), and Scientific & Technical (S&T). While the names may have changed since I left any Intel or Intel related assignments, the functions are still there and have been there since Wild Bill Donovan created the OSS. Unlike his British mentors who had 2 separate and distinct operational components - SIS (MI6) and SOE - Donovan had both clandestine collection and paramilitary ops under the DO function. When CIA came along, it adopted the same organization.

In my understanding of intelligence, the questions that policy makers need answered should drive the intel agency to frame requirements. this framing of requirements is inherently an analytical function and, therefore, belongs in DI. The analysts should be tasking the collectors with specific requirements while the collectors should be sharing even the seredipitous collections with the analysts which, in turn, should generate both new requirements and new questions from the policy makers. As far as the covert ops and paramilitary ops go, these are NOT inherently intel functions. Those who conduct them are intel consumers just as divisions, brigades, and battalions are.

But, you might ask, what about the cav squadron in a division? Is it not an asset of the G2? Although the 2 has tasking authority, the squadron commander is primarily responsible to the G3. Historically, rewards in cIA have gone to the covert operators who were involved in both clandestine collection and covert ops. The latter held the bigger rewards so th former got short shrift. (Jones may be making a reference to this phenomenon.) From my perspective, it appears that the CIA was run as if the Cav squadron were driving the train - deciding on its own authortiy whether to meet the G2 requirements or not because combat was more fun.

I was told by a DI guy seconded to DO that he was not allowed to share a critical report with his fellow DI analysts. I was told by a DO guy that my perception of many of the case officers in Latin America as "cowboys" was shared by case officers from other areas of operations (this was in the late 80s and early 90s).

So, does creating DNI just add an unnecessary layer of management? I don't think so because the DCI was too closely identified with the CIA to effectively manage the entire intel community. I was encouraged that the first DNI, John Negroponte, came from the intel consumer community and not from the producers (or even the analysts). As an aside, the only CIA director ever to come out of DI was SECDEF Bob Gates. All the rest have been case officers or intel managers (often from NSA like Gen Hayden and DNI Adm McConnell).

Bottom line: The solution is not in reforming one agency but rather reforming the entire community as well as the agencies that make it up. If I were King (it's good to be the King ) I would always have a consumer as DNI and an analyst as Dir CIA. I would split off covert ops and paramilitary from CIA - where to put them is an open question but they should be neither primary collectors nor their own analysts.

Cheers

JohnT