Space transportation company SpaceX provided this month what was a spectacle in two senses. One was the physical show of launching the most powerful rocket ever built. The other was how the whole affair, which ended with an explosion just four minutes into what was programmed to be an orbital mission, was described as a success, with congratulations from government officials and cheering by the company’s employees.
This way of defining success and failure reflects SpaceX’s engineering strategy, which involves launching a series of test vehicles with the expectation that each vehicle probably will have something wrong but will provide a learning experience to guide modifications on later versions. The upper stage of the Starship rocket, which never separated from its booster during this month’s launch, had already compiled a record of successive fiery crashes during earlier test flights.
This strategy is much different from the one that the government’s space agency, NASA, has had to follow. A failure of a NASA mission is regarded as a failure, period, and is not praised as a stepping-stone to some future hoped-for success. Because of that, the engineering that goes into a NASA mission is a more meticulous and time-consuming process aimed at achieving success in the fullest sense of the word every time a rocket is launched.
When NASA ignited its Space Launch System (SLS) rocket last November—making it at the time the most powerful rocket ever launched—it was only after numerous delays, which have been typical of NASA missions, as engineers checked every possible point of failure with the intention of making everything go right the first time. That first use of the SLS, in what NASA designated as Mission One of the Artemis program, was a success in the full sense, sending a spacecraft looping around the moon on a twenty-five-day mission.
This difference in the methodology of these two space programs is indicative of an expectation of perfection that often gets applied to government but not to the private sector. Some of the government agencies concerned are, like NASA, doing something as difficult as rocket science. Some routinely must address problems in which there are big information gaps—such as the intelligence agencies, that get looked to most on the very problems on which the information gaps are the biggest. Despite the inherent challenges involved—including the determination of adversaries to keep secrets and the unpredictability of many future events—when the intelligence community is unable to fill one of those gaps correctly on a matter that for some reason gets heightened public attention, this gets described as an “intelligence failure” amid calls for the problem to be “fixed.”
Understanding of such inherent challenges often is not extended to governmental missions but routinely is to all manner of activities in the private sector. In baseball, batters do not know what pitch the pitcher will deliver, and even the best batters in the major leagues fail to get a hit two-thirds of the time. But they do not get condemned for each out as a “batting failure.”
The chief reason for the differential treatment is that government programs are subject to politics, and politics involves incentives to find fault and demand accountability, regardless of the inherent challenges of a mission and the impossibility of achieving perfection. Those incentives are part of the process of one-upping political opponents and of politicians making names for themselves. The epitome of the process is the public congressional hearing in which committee members highlight in front of television cameras the less-than-perfect performances of governmental departments and demand changes. Daniel Dumbacher, a former NASA official who heads the American Institute of Aeronautics and Astronautics, said of SpaceX’s explode-and-learn method, “Government programs are not allowed to operate that way because … we have all the stakeholders being able to watch over and tell you no.”
The intense partisanship of current American politics intensifies this process. The motivation is strong to highlight any failings that can be associated with the other party, even if it is only some problem that falls in the area of responsibility of an executive branch agency, and that branch happens to be headed by a president of the other party.
Politics may underlie differential treatment between NASA and SpaceX regarding launch-related measures on the ground. At NASA’s Kennedy Space Center in Florida, the launch pads used for the largest rockets have flame trenches and water deluge systems that mitigate the effects of a blast-off. In SpaceX’s move-fast-and-break-things approach, no such infrastructure was constructed at its facility in Texas. The huge Starship rocket took off from a platform atop one corner of an ordinary-looking concrete slab.
As a result, effects on the ground were substantial, extending well beyond the SpaceX facility itself. The launch blasted a crater into the slab and threw debris more than a half mile away, onto a public beach, adjoining wetlands, and the ocean. A road to the beach remained closed until chunks of concrete and rebar could be cleared away. An enormous cloud of dirt and dust coated houses and cars in the town of Port Isabel, miles to the north. Some local activists have expressed concerns but there has not been a critical response from officialdom. Had this been a NASA operation—with a Democrat as head of the executive branch in Washington and Republicans in control in Texas—it is likely the official response would have been different.
Zero failures may sound like a beneficially aspirational, even if not practically achievable, standard to apply to government programs, but the application has costs. One is that it may simply not be the best approach for tackling large problems and achieving major goals. SpaceX’s own experience is suggestive. Although the company had numerous early failures as it was developing smaller rockets, such experimentation eventually led to the Falcon 9, which is now a profitable and reliable workhorse rocket for orbital missions. Imposing a less flexible standard on government programs may help to make self-fulfilling any argument that such programs are inherently less effective than risk-taking counterpart efforts in the private sector.
Another potential cost is that fixation on a failure and insistence on fixing it may introduce new problems in the fix. This is true, as I have described at length elsewhere, of much of the intelligence “reform” enacted after the 9/11 terrorist attacks, which has been established in popular discussion as a landmark “intelligence failure.”
The inordinate focus on a happening that the public and political class can easily and immediately brand as a failure may obscure larger issues at stake that by their nature are not so easily branded. A leader’s effort to avoid the easily identified type of failure may lead to policies that inflict greater costs on the nation than the failure would.
A leading example is the war in Afghanistan. The messy end in August 2021 to the U.S. involvement in the war has been repeatedly and vigorously labeled as a failure, especially by political opponents of President Joe Biden, who ordered that final withdrawal. But the withdrawal was a necessary pulling of the plug on a two-decade military expedition that had become a feckless effort at nationbuilding and entailed far more costs than anything incurred during the few days of denouement. The very swiftness of the collapse of the Ashraf Ghani regime and its security forces demonstrated the fecklessness and how even the best-planned withdrawal would have looked ugly. Three previous U.S. administrations (including the Trump administration, which negotiated a withdrawal agreement but did not execute it) shied away from pulling the plug and risking exposure to denunciations of “failure,” and in so doing kept the war going indefinitely.
Another sort of failure that those earlier administrations wanted to avoid—bearing in mind the general awareness of the history of the 9/11 attack—was an anti-U.S. terrorist attack that had some connection, however tenuous, to Afghanistan, which would be blamed on any president who had earlier withdrawn U.S. troops from that country. The anxiety about avoiding that type of failure obscured the reality that the Afghan Taliban, who took over the country in August 2021, constitute an insular movement that does not do international terrorism and is an enemy of the Islamic State, or ISIS, which is the group with a presence in Afghanistan that would be most likely to perpetrate such terrorism. A reminder of this reality came this week with word that a Taliban operation killed a leader of the local ISIS element who had planned a bombing at the Kabul airport that killed thirteen U.S. service members during the August 2021 withdrawal.
That news did not interrupt the political game of pouncing on a single “failure” to the exclusion of broader realities. The Republican chairman of the House Foreign Affairs Committee, Michael McCaul of Texas, grudgingly acknowledged the desirability of the ISIS leader’s demise but said that “this doesn’t diminish the Biden administration’s culpability for the failures that led to the attack” at the airport.
The subject of terrorism in the post-9/11 era has been especially prone to a zero-failures mindset that has spawned avoidable costs that exceed the potential harm of the feared terrorism itself. Those costs have included not only the material and human costs of foreign wars such as the one in Afghanistan but also encroachments on personal liberty and the moral stain of resorting to torture.