Jane Galt is a new addition to the links on your left. She has a series of posts on the proposed war on Iraq, all of which are well written and astute. This one, however, causes me to quibble. She states that the nuclear balance of power from the 1950s to the 1990s, which was based on the theory of mutually assured destruction (MAD, which has got to be the most appropriate acronym in the history of mankind) was stable.

Well, yes and no.

It was certainly very stable at the beginning of that time period and remained very stable until the 1970s or early 1980s. At that point, missle guidance technology had advanced to the point where the likelihood of a missile landing within a few feet of its intended target was very high.

There are three points in time when a nuclear power can launch its missiles:

Before the other guy does.

After the other guy does, but before his missiles start exploding overhead.

After the other guy does and after his missiles have already exploded.

The first option is a pre-emptive strike and, as far as I know, was never seriously considered as an option for US policy.

The third option is what our policy was from the beginning of the nuclear age until the mid to late seventies or early eighties (I may be wrong on the time frame, here, but probably not by much). This option really is mutual assured destruction: You can kill our nation, but not our retaliatory capacity. It will survive your strike, it will be launched following your strike and you will not survive that counterstrike. The same message is received from the adversary. Both messages are accepted as true. Ergo, no nuclear strikes.

But for that to work, our counterstrike capacity had to survive a first strike. With the improved guidance systems, this was becoming less and less likely. Fortunately, this was only true of our land based missiles. If our bomber fleet could get off the ground, the individual bombers became immune to nuclear strikes, and only a few of them needed to be able to penetrate Soviet and/or Chinese air defenses to inflict what was assumed to be unacceptable losses on the adversary. The other leg of our nuclear triad, missile submarines, (the boomers of Tom Clancy fame) were relatively immune at sea. The problem in both cases was communications with the "launch platform" after suffering a first strike.

We tried to find a way around the problem. For example each of our new MX missiles was to be housed in one of several silos and moved periodically. Playing peekaboo with our retaliatory capacity (the assumed target of any first strike) was designed to make a successful first strike harder by increasing the number of potential targets. Either the MX was never deployed (which is what I seem to recall) or MIRV technology (multiple independently targetable re-entry vehicles) negated the needle in the haystack advantage. In any event, in the seventies or eighties, it became stated US policy that its land based missiles would be launched when we received warning that either of our nuclear adversaries (PRC and USSR) had launched their missiles.

We literally had our fingers on a hair trigger for the better part of fifteen years. It still worked, but it wasn't what immediately comes to mind when the word "stable" is used.