In this Dec. 8, 1987, file photo, President Ronald Reagan, right, shakes hands with Soviet leader Mikhail Gorbachev after the two leaders signed the Intermediate Range Nuclear Forces Treaty to eliminate intermediate-range missiles during a ceremony in the White House East Room in Washington.

By February next year, the United States must decide whether or not to extend, for up to five years, the New START pact with Russia that regulates the two superpowers’ strategic nuclear arms. The treaty has been in force for almost 10 years. Before that, the original START pact was in force for over 15 years.

One of the Trump administration’s three conditions for extending the treaty is stronger verification. This is critical, given Russia’s history as a serial arms control violator. There is a strong case for “going back to the future” with the original START model, which provided for much stronger verification than New START.

To intuitively understand why this is, let’s imagine two gunslingers who want to work out a deal to keep their firepower in check. One is Wyatt Earp, deadly lawman. The other is his equally deadly arch foe, Ike Clanton, a known cattle rustler and horse thief, whom Wyatt would later face in the shootout at the O.K. Corral. Let’s compare how these two would structure their agreement under the original START approach versus the New START pact.

Revolvers here equate to missiles, caliber to missile “throw-weight” and bullets to warheads. Missiles, however, are finicky revolvers. You must test fire them with the number of warheads you intend to use in order to have confidence they really work.

The original START pact limited overall revolver firepower. It capped the gunslingers’ total caliber of all revolvers. It also limited the total number of all chambers in the revolvers. To make counting bullets simple and readily verifiable, it assumed that each chamber of every revolver is loaded with a bullet.

By contrast, the New START approach has no limit whatsoever on revolver caliber and only limits bullets actually loaded in the chamber. Neither does it have any limit on the number of chambers your revolver can have — only on the total number of revolvers. Both parties are even permitted to test fire revolvers with, say, 10 chambers, but have it counted with only one bullet, if that’s all it claims is loaded.

To verify the bullet limit under START, Wyatt obtains good confidence in knowing the caliber and chamber capacity of his rival by remotely observing (using satellites) his firing activity at his test range. He gets even better confidence because his rival must also turn over recordings of the dozens of test firings he conducts each year.

However, the New START verification approach has to be completely different because you have to confirm the number of bullets actually loaded in many concealed chambers. You are allowed to ask your rival to open up the chamber on a 2-3 percent sample of his revolvers each year to inspect them. You want to see if they are loaded with the same number of bullets he claims they are prior to your inspection.

Wyatt has big problems with the New START approach. First, in past agreements, Ike has obstructed and delayed inspections on which Wyatt relies to verify the deal. Worse, Wyatt realizes that even if he finds more bullets in Ike’s chamber than the number he claims in advance, he cannot logically conclude from this that Ike’s total number of bullets exceeds his limits. In other words, Wyatt cannot verify this bullet deal.

Wyatt’s New START problems aren’t over. It dawns on him that if tensions mount and a gunfight looms, Ike can simply, quickly and, perhaps without Wyatt knowing it, load up his extra chambers with bullets. Wyatt will have to upload in self-defense, and he starts to question whether this deal is better than no deal at all.

Wyatt’s verification problems are precisely those the United States faces under the New START pact. We cannot effectively verify the central warhead limits, even though two administrations have certified overall Russian compliance. However, such compliance conclusions are necessarily based on samples and significant assumptions and extrapolations. Further, the New START pact does not limit throw-weight and permits massive breakout capability in a crisis.

New START’s former chief U.S. negotiator has recently written that limiting “actual” warhead loading was “the most important innovation in New START.” However, it was arguably also the worst innovation because it gave rise to these central verification and breakout flaws. This same article also points out correctly that “[effective] verification regimes must not tempt either side to try an illicit treaty breakout.” But that is exactly what New START does.

In fact, identifying potential Russian treaty breakout activities was a top monitoring concern for the U.S. intelligence community. Given this breakout potential, the intelligence community discounted likely Russian cheating on the treaty’s unverifiable warhead limits. It would probably be more attractive for Russia to legally prepare to quickly and easily break out of New START constraints through warhead “uploading.”

There is yet another central verification and breakout problem with New START, which requires no metaphor to understand. This treaty, unlike the original START pact, does not place limits on non-deployed mobile missiles. Mobile missiles are much harder to detect and count via satellites than silo-based intercontinental ballistic missiles or submarine-launched missiles. Also, mobile missiles can be readily reloaded in war because they are very difficult to target.

Recognizing this, the START pact limited all mobile missiles. It also required, as did the Intermediate-Range Nuclear Forces Treaty, a continuous, on-site “perimeter portal monitoring” of the mobile missile final assembly plants. (Think of PPM as a kind of missile penitentiary, where missiles can only emerge at a single portal where they are thoroughly inspected and counted.)

New START, however, dispensed with all this — no mobile, non-deployed missile limits; no continuous PPM; and thus no effective verification of mobile missiles. But why?

According to the treaty’s chief negotiator, it was to save money by eliminating the “expensive” PPM program. Expensive? According to defense testimony, it cost $9.2 million annually to operate PPM in Russia under the INF Treaty. That’s half what Sam Houston State and Prairie View A&M colleges each spend yearly on their football teams.

Whether or not New START is extended, it must not serve as the model for future arms control verification. Rather, the United States should insist on a higher standard for effective verification in any future deal, and it should base verification on its solid ancestors — the INF Treaty and START.

Otherwise, in its deadly standoff with Russia, the United States, like Wyatt Earp in our parable, will have no confidence that it is actually limiting the firepower of its untrustworthy foe.

Bryan Smith is a senior fellow at George Mason University’s National Security Institute. He drafted the minority report on the New START agreement for the Senate Intelligence Committee and served as a verification expert on the INF Treaty and START in the U.S. Arms Control and Disarmament Agency.

More In Commentary