Some decades ago already the phrase “CNN effect” was coined, referring to a concern that the emotional power of pictures might generate such popular sentiment as to overwhelm the calm and careful deliberations of governments during crisis decision points nested in humanitarian debacles. The concern actually predated the name, since CNN set up shop only in 1980: It arose around the time of the 1968 Tet Offensive. It then reappeared as the “CNN effect” early in the post-Cold War world in a new context: Would U.S. foreign policy as “armed” social work, in other words, as a meliorist substitution for a policy rooted in geopolitical strategy—since that seemed to many to have been rendered obsolete—blossom in a unipolar world?

Some hoped it would; others feared it would. Either way, scholars have never pinned down an exact definition of the CNN effect, nor agreed on which Western 1990s-era interventions took place partly or largely on its account. Somalia? Bosnia? Kosovo? Was there no intervention in Rwanda because pictures were very much scarcer? Was public arousal or the lack thereof pretty much all there was to these decisions?

I doubt it. As best I can tell, the fears—or hopes, depending on one’s tastes—never really materialized. Even if attentive publics were more than usually spun up by vivid scenes of televised suffering, and well-situated pieces of such publics actually did things as a result—example: Bob Geldof’s 1984 “Live Aid” concerts, proximately inspired by BBC TV reporting on famine in Ethiopia¹—governments tended to remain circumspect and conservative when it came to putting their own soldiers in harm’s way. Over time, too, the novelty of seeing scenes of heartbreaking depredations on one’s TV screen seems to have faded; many people apparently became inured to it over time out of psychological self-defense. Then there is the more recent fracturing of the media landscape, and the complex cynicism about the veracity of news that has arisen partly as a result.

For all these reasons, it did not seem strange for the Atlantic to publish an article dated March 1, by staff writer Uri Friedman, titled “The ‘CNN Effect’ Dies in Syria.” Friedman wrote:

In recent days, the Syrian government’s relentless bombardment of the besieged rebel enclave of Eastern Ghouta has demonstrated just how muted the CNN effect is in Syria. We read that hundreds of people have perished in what the UN secretary-general describes as “hell on earth.” We see images of bloodied children, covered in rubble or shrouds, pop up in our Twitter feeds. We watch a doctor fall to the floor in tears because she knows she can’t save the life of a boy whisked to her overwhelmed hospital. “I was just making bread for him when the roof fell in,” the boy’s mother wails. “At least in heaven there’s food.” And yet there has been little public outcry in the United States over the military offensive and only belated and half-hearted efforts by world powers to stop it. News of the slaughter in Ghouta seems to be all around us, even as news of serious efforts to end it is nowhere to be found.

Yes, to claim that the CNN effect has died in Syria made sense a mere six weeks ago, except for one thing: Donald J. Trump.

Donald Trump is the CNN effect personified. He may be a lone atavistic emanation of it, but he is what he is and as President he is a lot. He proved the point in April 2017, when he claimed that seeing video of dead children gassed at Khan Sheikhoun led him to order a strike against Syria, and he proved it again this past weekend. Clearly, he is outrageable.

The problem here is twofold: first, that his capacity for outrage is directly proportional to what pictures he happens to see, which renders his outrage maddeningly episodic and hence unpredictable; and second that he shows no evidence of being able to bring to bear any other mental capability—analytical thought, for example—on which to base a decision beyond a spur-of-the-moment reaction to an imbibed “shot” of strong video.

As it happens, President Trump does not—indeed, cannot—act entirely alone in ordering the use of force: Thank God. So when he makes a decision to attack Syria on the basis of a impulse whose duration can be counted in a few hundred nanoseconds, various other warm bodies—a Secretary of Defense, a Secretary of State (if there happens to be one confirmed and on speaking terms with him), a National Security Advisor (the latest one of three hired “remotely” from Fox News broadcasts and in his job for less than a week), a press secretary, and others—will be milling about to cross “t”s and dot “i”s, so to speak. What targets should be hit, and why? Should the U.S. government act alone or with allies? How should the attacks be characterized for the world’s ear? What might go wrong, and what plans should be in readiness if something does go wrong—if, say, despite efforts at deconfliction the attack kills a lot of or the wrong kind of Russians?

These assorted warm bodies did a pretty good job by most accounts both last time and this time around. Last time, in April 2017, the 59 Tomahawks fired at Syria destroyed about 20 percent of the Syrian Air Force, with no U.S. loses of men or materiel, and no mistakes that would have spiraled us into escalatory trouble. This time, with very specific reliable intelligence that itself should send a powerful signal to the Syrian regime and its allies, the attack took out its targets cleanly and precisely, and arguably degraded the regime’s capabilities with regard to chemical weapons fabrication and hence future use.

The fact that the U.S. government acted with close allies represents a major positive symbolic statement, as useful within Europe as anywhere. After the effective collective response to the Skripal affair, it tells the Russians that their wedges are not thick enough to affect NATO in any significant way; and it dilutes the “America first” rhetorical toxins that have undermined allied confidence in American constancy. That is all to the good.

But where does this leave us? It leaves us, as the scientific-technical term would have it, completely ass-backwards. Major foreign policy decisions involving the use of force are supposed to be made after a process of analysis, deliberation, and planning; what is done with force is integrally bound up with how one plans to do it. Not in the Trump Administration, where first the President, in a “thought” process completely devoid of strategic focus, makes a decision to use force based on an emotional extrusion, and then his minders figure out how to implement the decision in the least damaging and dangerous manner. Even afterwards they can’t stop the President from screwing up the messaging: He can tweet “Mission Accomplished,” or call Bashar al-Assad an “animal” or a “monster,” or accuse Vladimir Putin of being complicit in the murder of children, without consulting anyone. That’s a problem. But most of the rest the warm bodies can shape, and at least in this case they have done so adeptly.

Alas, what the revolving cadre of warm bodies cannot do is create a strategy within which the use of force makes any real sense, even at less than “grand” levels, that is safe from presidential tampering. For example, Secretary of Defense James Mattis tried his hand at mid-level strategy in Syria several weeks ago when he suggested publicly that the modest American ground presence in Syria would remain and possibly expand. He had a handful of good reasons for saying such a thing: to disabuse the Syrian regime and others that it could reliably know the future of U.S. interests and investments in the area; to keep pressure on the territorial remnant of ISIS while America’s Kurdish allies’ attention and energy were distracted by Turkish military incursions elsewhere along the border; and above all to maintain a base for the expansion of the U.S. presence against the day when “skin in the game” might translate into leverage in a negotiation over the future of Syria. But before Secretary Mattis could say “don’t call me Mad Dog,” the President declared that he wanted U.S. forces out of Syria “very soon,” and pulled $200 million appropriated for stabilization and reconstruction in the area of the country under U.S. purview.

That, of course, would have essentially reestablished the conditions that called forth ISIS’s brand of Sunni self-help in the first place. Apparently, someone explained this to the President, and his determination to end the U.S. ground presence fell under a cloud of ambiguity as suddenly as it had appeared in bright light. There it remains, for all anyone knows.

This sort of thing, of which many examples can be cited, gives the Administration’s policy process the appearance of a 1952 Studebaker Starlight coup: the famous “coming and going” Stude whose front and rear appearance was so similar that you had to look twice to figure out if the vehicle was moving forward or in reverse. It was an amusing car; it’s not so amusing as a foreign policy optic.

President Trump’s decisions to use force in Syria are tethered to no strategy or even discernable purpose. They exist in their own time-and-space bubble, rather like unlinked episodes of “reality TV” shows. So what can anyone, no matter how experienced and well placed, make of them in terms of a strategy or a purpose?

Well, one purpose for such attacks, articulated best by British Prime Minister Teresa May and Foreign Minister Boris Johnson, is to uphold the taboo against the use of chemical weapons, by any power, anywhere. That is a valid purpose, not just for supporting international law as regards chemical weapons, but indirectly also for the status of other international arms control-related agreements: the NPT, the Outer Space Treaty, the Antarctica Treaty, and a few lesser-known others. But this argument really has little or nothing to do with Syria, and it is anyway a bit ethereal as purposes go, since the real glue holding these treaties together is the absence of any strong great power motive to violate them—yet.

The main purpose of this past weekend’s attack would presumably be deterrence: to get the Syrian regime to stop gassing its own people. That would be nice if it could be established, but even if it could be established it would not affect the course of the war or reduce significantly the regime’s capacity to murder and terrorize its own people by other means. Fewer than 1 percent of the 400,000-plus Syrian civilians murdered by their own government since 2011 have been killed with poison gas. The April 2017 attack didn’t deter the Syrian regime. It used chlorine gas on multiple occasions thereafter, presumably because chlorine is not one of the treaty-proscribed substances that fall under the aegis of the Organization for the Prohibition of Chemical Weapons (OPCW). The Douma attack apparently used chlorine mixed with some nerve agent—probably Sarin or VX judging from afar by the description of victims’ symptoms. That fits past patterns: the regime’s tactics advance incrementally, mounting more brazen attacks only after getting away with marginally less cruel ones.

The problems here multiply upon themselves; three, in particular, crawl all over each another. First, deterrence to the use of chemical weapons cannot be established by annual pin-prick scale attacks. U.S. leaders, preferably with but if not without its allies, would have to be prepared to hit the Syrians every time they crossed the line. That means a campaign, not a once-yearly gesture composed of cruise missiles. That’s a problem because of costs, given the still-parlous state of the Navy and the Air Force, and because of escalatory potential.

But, second, is it wise in any event to say, in effect, to the Syrian regime—and by implication to the Burmese government, for example, among others—that we don’t care if you terrorize segments of your own population and make hundreds of thousands of them into refugees as long as you don’t use chemical weapons to do it? How will future generations, not to speak of sentient present ones, size up that moral logic?

Besides, third, the Chemical Weapons Treaty, which entered into force in its current form only in April 1997, was designed and has always been understood since its origin in 1980 to apply to international conflict, not intrastate conflict. Perhaps it was self-servingly sophistic, but U.S. government lawyers pointed to that fact in excusing the U.S. and other Western governments for doing nothing when the Iraqi regime gassed its own Kurdish citizens in the Anfal Campaign of 1988, in Halabja most famously, but elsewhere as well. In a strict sense, Syria is no different in being an intrastate and not an international problem case.

But of course Syria today is different from Iraq in 1988. The main difference is that the context of the Iraq case was barely international; it can only be understood in the context of the final stages of the Iran-Iraq War, and lacked altogether a superpower dimension as Ba’athi Iraq’s Soviet ally plummeted toward blessed oblivion. Syria today manifests a war trifecta of sorts: a civil war nestled in a regional proxy war, in turn increasingly shadowed by a revenant great power competition (that is not, just by the way, usefully described as a new Cold War). This means that nudging any level of this war with force will affect the other two levels, and that in turn demands from any responsible government a clear sense of purpose: What exactly is a use of force supposed to accomplish over all, given its many moving parts, and what price are we willing to pay, and what risks are we willing to take, to accomplish it?

Ever since the so-called Arab Spring sideswiped Syria and led its regime to react in gratuitously gruesome ways, it was always going to be a hard problem for the United States. No U.S. administration in 2011 would have wanted to “own” Syria by projecting U.S. military force once again into the heart of the Middle East, yet no indirect way of influencing events promised much policy torque in a place whose inherent strategic importance was at best middling. Nothing the Obama Administration could have done—or as it turned out, mostly managed to avoid doing—promised success. Had it instead chosen to do what it ended up refraining from doing, it might have reaped the whirlwind anyway—just a different one.

Indeed, prospective sins of omission and sins of commission were both all too abundant in Syria in 2011, 2012, and beyond—and they still are. But when difficulties grow, true leaders do not throw up their hands and wave the “too hard” flag of de facto surrender, or make outsized claims about dubious achievements (example: “[W]e struck a deal where we got 100 percent of the chemical weapons out,” said Secretary of State John Kerry on Meet the Press, July 20, 2014). No, true leaders up their game.

So assuming for a moment that Donald Trump is not President of the United States, and that the foreign policy decision-making process therefore does not resemble a demolition derby with secretaries and advisers smashing into each other as they clog the doors coming and going, what might a rational strategy involving use of force in Syria look like?

In this case it would be wise to resort first to principle of exclusion: What can’t it look like? It cannot, or at any rate should not, look like a means to decisively influence the course of the war on the ground. Six or five or even four years ago a muscular if succinct use of U.S. airpower in tandem with a serious effort to not just arm but shake into political coherence the Syrian rebel opposition might have ended both the fighting and the Assad regime. But to do that now is all but impossible without risking a wider war that might well pull into the vortex not only Iran and Saudi Arabia (already at crossed swords over Yemen and potentially elsewhere, like Bahrain); but also Lebanon and Israel; Turkey more deeply than it already is; and possibly, if still improbably, provoke a direct U.S.-Russian clash. The risk-reward ratio is very unfavorable to an all-out politically ambitious use of force.

If it would be unwise to attempt too much, it is also unappealing to do too little. To essentially abdicate what there is of the U.S. policy investment in Syria, after having just used force to the accompaniment of some heady language, is to essentially say to the Iranians and the Russians: “OK, you win”—this despite the fact that both the Russians and the Iranians are overstretched by any objective measure. Do that and every U.S. partner or mere transactional associate in the region will conclude that they are on their own, which in turn could lead to dangerous forms of self-help like the disastrous Saudi form of self-help in Yemen. It’s not clear that this baleful impression would stop in the Middle East either. Of course, a true “America firster” cumisolationist would not care; but then a true “America firster” cum isolationist resembles a three-year-old playing hide-and-go-seek: Close my eyes against that tree and you can’t see me.

And there is a third thing that a Syria strategy involving a use of force comparable to that of this past weekend cannot, or should not, imagine it can achieve: drive the war to a genuine negotiating table. If this language sounds familiar, it’s because it mimics the drone of many Democratic congressmen and associated chatterati over the past few days. They must say something at a time like this, of course, but why do they seem to always say something so vapid?

To say that U.S. policy should aim toward a negotiated settlement of the Syrian war is correct, but as stated in recent days the advice is so vague as to be mere high-sounding pabulum. It might still make sense to say this if there were any evidence that the other parties seek a negotiated settlement, as well, in which case the use of force would not even be critical to get a deal—and indeed, some Democrats have argued as much. But right now those other parties don’t seek a negotiated settlement, but rather seek to win militarily; and they think they can. The Syrian regime’s idea of a negotiated settlement mirrors exactly its idea of a truce with rebel fighters in pockets of resistance: It’s a euphemism for surrender, accompanied by a unenforceable promise that government forces won’t then kill your family anyway.

The Syrian regime has left to conquer only a small chunk of territory in the south near the Jordanian border and most of the province of Idlib in the country’s northwest. Idlib is where the next major bloodfall is likely to occur in the war, equaling if not excelling in savagery last year’s Aleppo campaign. As long as no foreign power is prepared to oppose it, directly or otherwise, the Iranians and the Russians will support the Syrian regime because the cost to them will be low. Ah, now we come to the nub.

The one realistic thing U.S. military power can do in Syria is raise the cost to the Russians and the Iranians of their support for the regime, in hopes that doing so will split the coalition and lead one party—the Russians, presumably, who are overstretched perhaps more than the Iranians—to quietly say “enough” to the other two. Raising Moscow’s costs was the minimal point of arming the Afghan mujahideen against the Red Army, remember, and it’s a plausible argument, all else equal, for providing weapons to the Ukrainian government. So this is not a new idea, just an application of an old and tested one.

How would we do this, now that we have long since given up on cultivating local proxy forces, and have all but alienated effective ones we once had among the Kurds? Not by directly attacking Russian assets or Iranian proxies, but by attacking elite Alawi units in the Syrian military. Loyal manpower is the regime’s weakest link. The Alawi leadership does not trust other Syrians in these positions, so the more of these clansmen that are put out of action, the more the Russians and Iranians (directly or through Hezbollah and assorted Shi’a militias) will have to substitute for them to keep the war tilted in their favor. They have already bled at the hands of rebel opponents, but they must be made to bleed a lot more, translating into higher expenses and more casualties for both, before they will reconsider their policies.

But an effort to raise costs and split the Assad-Iranian-Russian coalition cannot wait much longer. If Idlib falls, the non-salafi resistance will be too weak and scattered to matter, and hence no Syrian proto-political parties will exist that are capable of benefitting from and, more important, sustaining the compromise terms of a negotiated settlement. Outsiders, even Arab outsiders, cannot do this by themselves, after all. So if the Trump Administration were to use force purposefully, strategically, in Syria, this is how, and when, to do it.

Is this approach without risks? Of course not; at this stage in this tri-level war there are no effective risk-free options. In response to a U.S. effort to raise Russian and Iranian costs, the two targets could double down and prefer to pose a threat of escalation to the United States and any allies willing. So before embarking on such a course, we must think through possible scenarios and decide how far is too far to go chest bumping with bears and lions.

My sense is that the Russians are realistically risk-averse, and could be peeled off, especially if Washington were to express an openness to some reasonable diplomatic compromise over Ukraine. Getting the Iranians to then settle for part of a Syrian loaf in a genuine negotiated settlement would be tougher, but that is to be expected since the Iranian pretention to regional hegemony—not the future of the Syrian government or even the Russian position in the Eastern Mediterranean—is what should be focusing U.S. policy attention most these days. To the extent the Russians lose incentive for continuing in the role of military adjunct in Syria, that would dump the demurred Russian costs of supporting the Assad regime onto Iran. The United States and its allies would then be in a position to raise those costs, too.

This is the only realistic meaning of using force to shove the Syrian conflict toward the negotiating table. It would involve a good deal more force than was employed this past weekend, indeed, a level of force that all but a few Democrats (and not that many Republicans either) would be willing to countenance. So aside from the political challenge it poses, such a course of action would presage a daunting and uncertain task that requires clear-sighted leadership with both tenacity and patience to see it through. Those who are now babbling about this policy course need to understand what is really involved, and all of us need to take the measure of the emotionally volatile weathervane-headed Administration that would be in charge of it.

That is what Syrian “hard” looks like in the aftermath of this past weekend’s strike. It doesn’t look pretty.