Editor’s Note: Please also take a look at our recent piece If America Could Rebuild the U.S. Navy from Scratch.

“If I were creating the world I wouldn’t mess about with butterflies and daffodils. I would have started with lasers, eight o’clock, Day One.”

So said the personification of Evil in Terry Gilliam’s 1983 classic film, Time Bandits.  If we were able to dismantle the entire defense establishment and bring it back one piece at a time with 2014’s technology, what would it look like? In the case of the U.S. strategic nuclear deterrent, I would start with land-based intercontinental ballistic missiles buried in silos in North America.        

This might seem an odd choice. Why choose such destabilizing weapons? After all, ICBMs are ill-suited for very much besides killing millions of people. They only have an offensive role; they cannot be meaningfully defended; they cannot be recalled once launched; their existence cannot be hidden; their flight times are terrifyingly short. Why not start with something with a little more flexibility?

History and the state of technology in the 1950s did in fact lead to a different conclusion at the time. Before the invention of long-range missiles, the United States and the Soviet Union created huge bomber forces over six decades ago. All of the vices of missiles are reflected – supposedly – in the virtues of bombers: they’re slow, they can be recalled, they have human beings in them who can make real-time decisions, and they can be used for a broad array of missions. They are a prefect nuclear Swiss Army knife, suited for all kinds of missions.

That’s exactly what makes them dangerous, however. They’re tempting to policymakers who might want to use nuclear weapons while remaining in denial about the immense consequences of using nuclear weapons. If the goal is purely to deter the use of nuclear weapons against us, ICBMs (and their submarine launched little brothers, SLBMs) are adequate to the job. If the goal is to give the President all kinds of options for nuclear use – which I would seek to avoid – then by all means, buy more bombers. If the mission is purely deterrence, however, we have to rethink the accumulated decisions of a different time.

The notion of going toe-to-toe with the Russkies in nuclear combat made at least some sense in the 1950s. During the Cold War, we expected to fight the Soviet Union in a protracted nuclear war with far smaller nuclear weapons than we have now. The whole thing was supposed to look like World War II, only bigger: conventional war in Europe, naval engagements on the seas in and around Eurasia, fighter jocks once again blasting each other out of the German skies. In the midst of that war, bombers would not only have discrete missions against enemy military targets and infrastructure, they could be used for signaling, warnings, and “intra-war” deterrence (as in, “my bombers are in the air, and you now have eight hours to back off”).

ICBMs appeared at the end of the 1950s and were institutionalized as the backbone of the American strategic deterrent by 1970. Once missile technology appeared, the only real question about missiles was whether to hide them underwater, or to bury some in holes while attaching others to aircraft. The Air Force won this bitter interservice dispute, and Admiral Arleigh Burke and the Navy lost in their bid for a “minimum deterrent” of a few hundred missiles at sea. (The Air Force, Burke later complained, was as “ruthless as the Communists” and used the same methods.)

In the end, America split the difference three ways: henceforth, we would have a “triad” of forces, whose goal was less to create options than it was to survive a total Soviet onslaught with enough bombs left to reduce the USSR to ash. So why not just replicate this structure today? After all, that’s what we keep doing anyway. Every revision of our nuclear policies only ends up producing a Mini-Me of our previous force structures. We still have land, sea, and air based weapons, just in smaller numbers and tighter configurations. And it must have worked, since we’re all still here. Right?

Maybe. But the conditions that created the nuclear force of 1964 are not the conditions of 2014. If we’re serious about establishing nuclear weapons solely as a deterrent against nuclear attack on the U.S. or its closest allies, then we have to structure our forces so they can do nothing else but retaliate for such an attack.

ICBMs are the perfect retaliatory weapons precisely because they cannot be hidden, defended, or tailored to tactical missions. Moreover, they have one essential quality that establishes their deterrent value: they are located in the United States of America. There is no way to attack those weapons, and to strip America of its retaliatory capability, without attacking the sovereign territory of the United States. Prospective enemies must be denied any hope of half-way measures against us: if they mean nuclear war, then they must decide upon nuclear war.

Nuclear exchanges on the high seas or plucky bomber pilots making their runs deep inside enemy territory only happen in Tom Clancy’s pulpy novels. During the Cold War, we thought hard about how to convince the Soviets that a conventional war – one we would certainly lose on the ground – would escalate to nuclear conflict. Today, America is the preeminent conventional power and will remain so for decades to come (assuming we decide to maintain those abilities). There is nothing that nuclear weapons can do for us, short of deterring nuclear use by others, which we cannot achieve with conventional arms.

This issue of nuclear “utility” is one we avoided during the Cold War in the name of deterrence.  But sooner or later, we have to face two realities. First, the consequences of even a tiny nuclear exchange are more than any of us – and that includes Russia and China – are really willing to bear. Second, the use of even the smallest nuclear arms in crowded areas like Asia and the Middle East will inflict costs on our friends and allies that we, and they, will never accept.

So here’s what our new deterrent should look like.

America can defend itself with a small number of strategic nuclear arms, perhaps as low as two or three hundred. Scholars at the Air War College have pegged a number close to 300, which might even be high, but since we’re already at 1550 by treaty with Russia, there’s room to cut.  Start with a base of 300 bombs, and keep a third of those on single-warhead missiles in North America. (ICBMs with one warhead would be useless as first-strike weapons and thus their mission of retaliation would be clearer.) Assume the enemy surprises us – these days, pretty much a ridiculous assumption, but always plan for the worst – and somehow manages to destroy two-thirds of them before we even know what’s happening. That leaves 25 land-based missiles landing on enemy cities and infrastructure, killing millions, inflicting almost unrecoverable damage, and raising the cost of war beyond any benefit gained by this notional attack.

And then there are the submarines.

If four U.S. nuclear submarines with 16 to 20 missiles (and one to three warheads on each missile) are on patrol at any given moment, that means that the 100 warheads of the land-based force will always be backed up by anywhere from 64 to 240 warheads. Even at the lowest number, with three out of four lost to enemy action, one submarine would still be left to level the enemy’s capital and a dozen more cities. If that isn’t a deterrent, nothing is.

This reboot would streamline our nuclear forces, simplify our overly-complex nuclear strategies, and save billions of dollars. Conventionally-armed bombers will always be crucial in picking up missions that require the heavy use of airpower. (Bombers aren’t going away.) But nuclear bombers would be a “nice to have” not a “must-have,” and before arming a single bomber with a nuclear bomb, someone would have to make a case based on strategy rather than tradition.

When I worked for the late John Heinz in 1991 when he was the senior senator from Pennsylvania, he voted against continuing funding for the B-2 strategic bomber. The B-2 advocates came to our offices, and made no better case for the nuclear mission of the B-2 other than to say: it’s a cool weapon for fighting a long nuclear war with the Soviet Union…oh, and there are lot of subcontractors who need production to keep going. Heinz was undecided until the last minute, and then he went and voted. I was waiting for him outside the Senate chamber, and when came out, he said: “I did the conservative thing: I voted to save money.”

Start from zero, and add enough survivable nuclear weapons to destroy the regime of any nation that attacks us with nuclear arms. Put all the exotic toys away, focus on that one mission, and America will do not only a conservative thing, but a sensible thing, by saving money and confirming that our nuclear weapons actually have a real reason for existing besides the science-fiction scenarios of nuclear warfighting.