Kahn, a military analyst (with the Hudson or Hoover Institute, I believe) put together the first serious challenge to the prevailing doctrine of thermonuclear war, Mutually Assured Destruction (MAD). MAD, a policy which relied on deterring a nuclear aggressor by promising a devastating retaliation to a nuclear attack, was the strategy that had evolved during the early Cold War. No one was really happy with it, it had a "put all your eggs in one basket" feel to it, and it meant that in order to ensure it work properly it had to have a hair-trigger quality to lend it credibility and which troubled everyone. What happened in the event of an accidental attack, or a move by a rogue third party, or a renegade commander acting without orders.? Could a computer or clerical error start a war by mistake? In order to make the retaliatory force robust enough to survive the fog of a first strike, it had to be designed with a binary trigger, go or no-go. There was no way war could be stopped once it started, no way to negotiate your way out of a tragic, but limited, initial exchange. It was apparently an extremely unstable stand-off, or it appeared so at the time. History showed otherwise, but Kahn's criticisms of MAD were nonetheless valid and they struck a nerve with policy makers on both the left and right. No one knew if MAD would work, it couldn't be tested under real world conditions and we had no experience with this type of situation, we had no idea whether it would work or if it would work too well. Critics even pointed out a major flaw, a full response to an enemy nuclear attack would not save the attacked party, it would simply destroy the world out of spite. Could a cynical enemy take advantage (rightly or wrongly) of his belief that his devastated enemy might decide there was really no point in returning fire, or might want to but not be able to, or might just lose his nerve at the last minute? Kahn brought up a lot of points people were already well aware of, but didn't like to talk about.

Kahn's solution to this dilemma was not unreasonable, and it should not have been dismissed out of hand. And his case was very well thought our and cogently argued. His position was that in order to remove the mad inevitability of MAD, nuclear wars had to be fought like other wars, with an eye to unanticipated developments and with command and control capable of exercising options for dealing with the shadow zone between total peace and all-out war. He reasoned a limited war might be an alternative to a full-scale war, and perhaps might even prevent the latter. He argued for flexibility in tactics and an evolving strategy. He suggested that nuclear wars could be fought, that they had winners and losers, and might even end in a negotiated stalemate acceptable to both sides. He made logical arguments that even the increased possibility of a limited nuclear exchange was justifiable if it made the certain probability of global suicide or total defeat less likely.

His program was multi-pronged, and it threw the conventional wisdom on it's ear. It involved keeping the retaliatory capability, but adding to it other elements, including strategies which could be interpreted as provocative. He argued that this was exactly what the Russians were doing anyway, and he was right. The Russians, always playing catch-up ball with us, were ready to try anthing that might give them an edge. Kahn's proposals included a US first-strike capability, anti-missile defenses, and a comprehensive civil defense program. He concluded that this would not only give the US multiple options in a war, capable of graduated escalation short of all-out exchange, but that it would provide a complication to the USSR's war planning and force them to dilute their strengths by compelling them to divert resources to weapons programs they might never use or even need. The idea was that we could afford an arms race better than they could.

Almost all of Kahn's suggestions were eventually adopted by American war planners, in some form, primarily because of our technological advantages over the Soviets. Little effort was made to upgrade or protect our second strike capability. Highly accurate silo-busting missiles, equipped with MIRVs were deployed, even on our nuclear submarines (supposedly a purely retaliatory weapons system). These were clearly a first-strike option. Elaborate technologies and strategies to frustrate Soviet second-strike capabilities were also developed, and widely publicized so they would know we were doing it.
Others were developed in secret so the Russkis wouldn't know until they launched their retaliation in response to our first strike. Trust me, I know this to be true, from personal experience.

Anti-ballistic missile defenses were proposed, in spite of ABM-banning treaties which had been insisted on by the US only a few years earlier because of their de-stabilizing consequences. These new weapon systems were developed at the expense of neglecting our retaliatory arsenal, as if we had no intention of ever having to rely on it too heavily. The only aspect of Kahn's vision which was not implemented was a massive government-administered Civil Defense effort, and I suspect that was because it was the one component not requiring massive spending programs primarily benefitting defense contractors; it would be a big government bureaucracy with little private sector participation. I should make it clear at this point that these decisions were not necessarily ideologically motivated, it's just the way the dynamic of money and power and votes works in Washington. This was the compromise that eventually emerged from the defense debates of the Cold War, and it was eventually voted in by both Democrats and Republicans.