Webinar

Academic Webinar: Military Strategy in the Contemporary World

Wednesday, October 25, 2023
Getty
Speaker

Adjunct Senior Fellow for Defense Policy, Council on Foreign Relations; Professor of International and Public Affairs, Columbia University

Presider

Vice President for National Program and Outreach, Council on Foreign Relations

Academic and Higher Education Webinars and National Security and Defense Program

Stephen Biddle, adjunct senior fellow for defense policy at CFR and professor of international and public affairs at Columbia University, leads the conversation on military strategy in the contemporary world.

FASKIANOS: Welcome to today’s session of the fall 2023 CFR Academic Webinar Series. I’m Irina Faskianos, vice president of the National Program and Outreach here at CFR.

Today’s discussion is on the record, and the video and transcript will be available on our website, CFR.org/academic, if you would like to share them with your colleagues or classmates. As always, CFR takes no institutional positions on matters of policy.

We’re delighted to have Stephen Biddle with us to discuss military strategy in the contemporary world. Dr. Biddle is an adjunct senior fellow for defense policy at CFR and professor of international and public affairs at Columbia University. Before joining Columbia he was professor of political science and international affairs at George Washington University. He regularly lectures at the U.S. Army War College and other military schools and has served on a variety of government advisory panels and analytical teams, testified before congressional committees on issues relating to the wars in Iraq, Afghanistan, and Syria; force planning; conventional net assessment; and European arms control, just to name a few. And, finally, Dr. Biddle is the author of numerous scholarly publications and several books, including his most recent, Nonstate Warfare, published by Princeton University in 2021 and he just recently authored a piece in CFR’s magazine Foreign Affairs in the September/October 2023 issue entitled “Back in the Trenches: Why New Technology Hasn’t Revolutionized Warfare in Ukraine,” and we shared that out in the background readings for this conversation.

So, Steve, thank you for being with us. I thought you could give us an overview of the changes you’ve seen in military operations as a result of technological innovation and say a few words about wartime military behavior especially as you’ve studied it over the years and what we’re seeing now in Ukraine and now with the Israel-Hamas war.

BIDDLE: Yeah, I’d be happy to. There’s a lot going on in the world of military affairs and strategy at the moment between Gaza, Taiwan Straits, and, of course, Ukraine.

Maybe as a conversation starter I’ll start with Ukraine but we can go in whatever direction the group wants to go in, and the spoiler alert is in the headline of the article from Foreign Affairs that you’ve already assigned.

There’s a big debate over what Ukraine means for the future of warfare and what Ukraine means for the way the United States should organize its military, modernize its equipment, write its doctrine and so on. One of the most common interpretations of what Ukraine means for all this is that it’s harboring—it’s a harbinger of a revolutionary transformation.

The new technology, drones, space-based surveillance, precision-guided weapons, hypersonics, networked information, artificial intelligence, this whole panoply of things in this argument is making the modern battlefield so lethal, so radically more lethal than the past is that in the present and in the future offensive maneuver will become impossible and we’ll get the dawn of some new age of defense dominance in conventional warfare, which, if true, would then have all sorts of implications for how the United States should make all these kinds of defense policy decisions.

As those of you who read the Foreign Affairs article know I don’t buy it because I don’t think the evidence is consistent with that supposition. You’ll be happy to hear that I’m not planning to do a dramatic reading of the Foreign Affairs essay, entertaining as I’m sure that would be, but I did think it might be useful for me to briefly outline the argument as a way of teeing up the subsequent conversation.

And the basic argument in the article is that whereas there are, indeed, all sorts of very new technologies in use in this war, when you actually look carefully at the results they’re producing, at the attrition rates that they’re actually causing, at the ability of the two sides to gain ground and to suffer the loss of ground, the actual results being produced by all this very new technology are surprisingly less new than is assumed and supposed in the argument that we’re looking at some transformational discontinuous moment in which a new age of defense dominance is dawning.

This doesn’t mean that nothing’s changing or that the United States military should do in the future exactly what it’s done in the past. But the nature of the change that I think we’re seeing is evolutionary and incremental as it has been for the last hundred years, and if you think what’s going on is incremental evolutionary change rather than discontinuous transformation that then has very different implications for what the U.S. should do militarily.

So just to unpack a little bit of that by way of pump priming let me just cite some of the examples of what one actually observes and the outcomes of the use of all these new technologies as we’ve seen in Ukraine.

So let’s start with casualty rates and attrition. At the heart of this argument that new technology is creating a new era of defense dominance is the argument that fires have made the battlefield so lethal now that the kind of offensive maneuver you saw in World War II or in 1967 or in 1991 is now impossible.

And, yet, the actual attrition rates of, for example, tanks, right—tanks tend to be the weapon system that gets the most attention in this context—are remarkably similar to what we saw in the world wars. So in the first twelve months of the fighting in Ukraine, depending on whose estimates you look at the Russians lost somewhere between about half and about 96 percent of their prewar tank fleet in twelve months of fighting.

The Ukrainians lost somewhat in excess of 50 percent of their prewar tank fleet, and intuitively that looks like a heavy loss rate, right? Fifty (percent) to 96 percent of what you opened the war with, that seems pretty—you know, pretty dangerous. But in historical context it’s actually lower than it frequently was in World War II.

In 1943, the German army suffered an attrition rate to the tanks it owned at the beginning of the year of 113 percent. They lost more tanks in 1943 than they owned in January 1943. Their casualty rate went up in 1944. They lost 122 percent of all the tanks they owned in January of 1944.

So these attrition rates while high aren’t unusually high by historical standards. What about artillery, right? Artillery is the single largest casualty inflicter on the modern battlefield defined as since the turn of the twentieth century, 1900.

As far as we can tell the attrition rate from Ukrainian artillery fire of Russian forces in this war looks to be on the order of about eight casualties inflicted per hundred rounds of artillery fired and that’s higher than in World War II but not discontinuously radically higher.

In World War II that figure would have been about three casualties per hundred rounds fired. In World War I that figure would have been about two casualties per hundred rounds fired. If you chart that over time what you see is an essentially linear straight line incremental increase over a hundred years of about an additional .05 casualties per hundred rounds fired per year over a century of combat experience. There’s no sudden discontinuous increase as a result of drones or networked information or space-based surveillance at the end of the period.

What about ground gain and ground loss? The purpose of attrition on a modern battlefield is to change who controls how much territory and the whole transformation argument is that all this putatively much more lethal technology is making ground gain much, much harder than in the past, and yet the Russia offensive that opened the war, mishandled as it was in so many ways, took over 42,000 square miles of Ukraine in the first couple of months of the war.

The Ukrainian Kyiv counteroffensive retook more than 19,000 square miles. Their Kharkiv counteroffensive retook 2,300 square miles. The Kharkiv counteroffensive took back more than 200 square miles. There’s been plenty of defensive stalemate in the war, right? The Russian offensive on Bakhmut took ten months to take the city. Cost them probably sixty (thousand) to a hundred thousand casualties to do it. The Mariupol offensive took three months to take the city.

But this war has not been a simple story of technologically determined offensive frustration. There have been offensives that have succeeded and offensives that have failed with essentially the same equipment. Drones didn’t get introduced into the war in the last six months. Drones were in heavy use from the very outset of the fighting and this kind of pattern of some offensives that succeed, some offensives that don’t, like the attrition rate is not particularly new.

I mean, the popular imagination tends to see World War I as a trench stalemate created by the new technology of artillery and machine guns and barbed wire and World War II as a world offensive maneuver created by the new technologies of the tank, the airplane, the radio. Neither World War I nor World War II were homogeneous experiences where everything was defensive frustration of World War I and everything was offensive success in World War II.

That wasn’t the case in either of the two world wars. The Germans advanced almost to the doorsteps of Paris in the initial war opening offensive in 1914. In 1918, the German spring offenses broke clean through Allied lines three times in a row and produced a general advance by the Allies and the subsequent counteroffensive on a hundred-eighty-mile front. There was a lot of ground that changed hands in World War I as a result of offensives in addition to the great defensive trench stalemate of 1915 to mid-1917.

In World War II some of the most famous offensive failures in military history were tank-heavy attacks in 1943 and 1944. The Battle of Kursk on the Russian front cost the German attackers more than a hundred and sixty thousand casualties and more than seven hundred lost tanks. The most tank-intensive offensive in the history of war, the British attack at Operation Goodwood in 1944, cost the British a third of all the British armor on the continent of Europe in just three days of fighting.

So what we’ve seen in observed military experience over a hundred years of frequent observational opportunity is a mix of offensive success and defensive success with technologies that are sometimes described as defense dominant and, yet, nonetheless, see breakthroughs and technologies that are sometimes seen as offense dominant and, yet, sometimes produce defensive stalemates and what really varies is not so much driven by the equipment, it’s driven by the way people use it.

And the central problem in all of this is that military outcomes are not technologically determined. The effects of technology in war are powerfully mediated by how human organizations use them and there are big variations in the way human organizations use equipment.

And if you just look at the equipment alone and expect that that’s going to tell you what the result of combat is going to be and you don’t systematically account for how the human organizations involved adapt to what the technology might do on the proving ground to reduce what it can do on the battlefield then you get radically wrong answers and I would argue that’s what’s going on in Ukraine.

Both sides are adapting rapidly and the nature of the adaptations that we’re seeing in Ukraine are very similar to the nature of the adaptations we’ve seen in previous great power warfare. Again, incremental lineal extensions of emphases on cover, emphases on concealment, combined arms, defensive depth, mobile reserve withholds—these are the ways that all great power militaries have responded to increasingly lethal equipment over time to reduce their exposure to the nominal proving ground lethality of weapons in actual practice.

The problem is this collection of techniques—and in other work I’ve referred to them as the modern system, this kind of transnational epistemic community of practice and the conduct of conventional warfare—to do all these things right and minimize your exposure is technically very challenging. Some military organizations can manage this very complex way of fighting; others cannot. Some can do it on one front and not on another front, and the result is we get a lot of variance in the degree to which any given military at any given moment embraces the entirety of this doctrinal program. Where they do, defenses have been very hard to break through for a hundred years. This isn’t something that came about in February of 2022 because of drones and networked information. This has been the case repeatedly for a century of actual combat.

But where they don’t, where defenses are shallow, where reserve withholds are too small, where combined arms aren’t exploited, where cover and concealment isn’t exploited, then casualty rates go way, way up. Then breakthrough becomes possible. Then attackers can gain a lot of ground with tanks or without tanks.

The German offensives that broke clean through Allied defensive lines in 1918 had almost no tanks. The first of them, Operation Michael, was a one-million soldier offensive that had exactly nine tanks in support of it.

So the differences that have mattered are the interaction of increasingly lethal technology with these variations and the ability of real human organizations to master the complexity needed to fight in a way that reduces exposure to this and that’s the same thing we’ve seen in Ukraine.

Where defenses have been shallow and haven’t had enough reserves behind them you’ve gotten breakthroughs. Where they’ve been deep, adequately backed by reserves, as we’ve seen in this summer counteroffensive over the last three or four months, for example, they’ve not been able to break through and this isn’t a new story. This is just a recapitulation of a hundred years’ worth of military experience.

If that’s so then what difference does it make to the U.S.? So, again, as I suggested earlier, that doesn’t mean don’t change anything, right? A 1916 tank on a modern battlefield would not fare well. Part of the stability in these kinds of outcomes is because people change the way they do business. They change the way they fight. They update their equipment. They execute measure/countermeasure races and so we need to continue to do that.

Depth is probably going to increase. Reserve withhold requirements are going to go up. Demands for cover and concealment are going to increase.

There will be technological implications stemming from the particular measure/countermeasure races that are emerging now especially with respect to drones. Almost certainly the U.S. Army is going to have an incentive, for example, to deploy counter drone escort vehicles as part of the combined arms mix, moving forward.

But the principle of combined arms that’s behind so much of the way the U.S. Army fights is very unlikely to change very much. What’s going to happen is a new element will be added to the combined arms mix, and escort jammers and anti-aircraft artillery and other air defense systems that are optimized for drones will become part of the mix of tanks and infantry and engineers and signals and air defense and all the rest, moving forward.

The whole revolution argument, though, is not that, right? The reason people refer to this as a revolution, as transformation, is they’re using language that’s designed to tee up the idea that ordinary orthodox incremental updating business as usual isn’t enough in this new era because of drones, because of hypersonics, or space-based surveillance or whatever.

We need something more than that, and I think if we look closely at what’s going on in Ukraine what we see is not an argument that we need to transform the way the U.S. military does business. What we see is an argument for incremental change that implies incremental adaptation is appropriate, that it’s not the wrong thing to do.

I think it’s possible to over-innovate. I think there are ample historical examples of militaries that have gone wrong not by being resistant to innovation—there are plenty of those, too—but by doing too much innovation. In the 1950s and 1960s U.S. Air Force transformed itself around an idea that conventional warfare is a thing of the past, all wars of the future will be nuclear, and they designed airplanes for nuclear weapon delivery that were horribly ill-suited to the conventional war in Vietnam that they then found themselves in.

The U.S. Army transformed its doctrine following a particular understanding of the lethality of precision-guided anti-tank weapons in the 1973 Arab-Israeli war, adopted a concept called active defense that relied on static defense in a shallow disposition from fixed positions, emphasizing the ostensible new firepower of anti-tank weapons. Found that that was very innovative but very ineffective and abandoned it in favor of the airline battle doctrine that’s a lineal descendant of the doctrine we use now, which was much more orthodox and conventional.

There are plenty of examples of militaries that have over-innovated. This language of revolution and transformation is designed to promote what I’m concerned could be over-innovation again. I think we could talk more about the particulars of what incremental adaptation should comprise but I think that’s the right way forward in light of what we actually observe about what’s going on in Ukraine.

FASKIANOS: Fantastic. Thank you for that, Steve. That was great.

Let’s go now to all of you for your questions.

(Gives queuing instructions.)

And so don’t be shy. This is your time. We have our first question from Terrence Kleven.

Q: Hello. Can you hear me?

FASKIANOS: We can. If you could tell us your affiliation that would be great.

Q: Yes, very good. Terrence Kleven. I’m at Central College in Pella, Iowa, and I teach in a philosophy and religious studies department and I teach quite a lot of Middle Eastern studies.

Thank you very much for your presentation because so much of this we don’t talk about enough and we don’t understand, and I appreciate the opportunity to hear what you have to say and look forward to reading your—some of your material.

Just kind of a practical question, why aren’t the Russians using more planes in this war or are they and we just don’t have a report of that? I assume that the Russian air force is much superior to what the Ukrainians have but it doesn’t seem to give them a great advantage. What’s missing? What’s going on?

BIDDLE: Yeah. You’re raising a question that has bedeviled military analysts in this war since its beginning. Part of the issue is the definition of what plane is, right? If we define a plane as something that uses aerodynamic lift to fly through the air and perform military missions the Russians are using lots of planes; they just don’t have pilots. We call them drones. But a drone, to a first approximation, is just a particular inexpensive, low-performance airplane that is relatively expendable because it’s inexpensive. But because it’s inexpensive it’s also low performance. If by airplanes one includes drones, then there’s lots of airplane use going on.

What you had in mind with the question, I’m sure, is the airplanes that have people in them—why aren’t they more salient in the military conduct of the war, and the Russians have tried to use piloted aircraft. The trouble is the loss rates have kept them, largely, out of the sky.

So this again gets back to the question of human adaptation to new technology. Air forces—and navies, by the way, but that’s a different conversation—are much more exposed to more technology increases—the technology changes that produce increasing lethality than ground armies are.

Ground armies have much easier access to cover and concealment. It’s hard to find much cover and concealment up there in the sky, right? You’re highlighted against a largely featureless background. There are things you can do as an air force to try and reduce your exposure to precision-guided anti-aircraft weapons and the U.S. Air Force, for example, practices those extensively.

But the complexity of operating an air force to be effective at the mission called SEAD—suppression of enemy air defenses—is very high and it requires a lot of practice and it requires a lot of flight hours and it requires you to burn a lot of fuel in training, and the U.S. Air Force is willing to do that. The Russians historically have not.

Therefore, they’re not very good at it. Therefore, they’re very—they have been very exposed to the lethality precision-guided Ukrainian anti-aircraft defenses and, therefore, they’ve mostly decided not to expose themselves to this fire.

They fly mostly over friendly terrain, especially in metropolitan Russia, and they fly at low altitudes that keep them under the radar, which is a cliché that’s leached into public conversation because of the actual physics of the way radar works and responds to the curvature of the earth.

If the Russians operate over Russian territory at low altitude and launch cruise missiles at huge distances then their airplanes don’t get shot down as much. But then the airplanes are a lot less effective and contribute a lot less and that’s the tradeoff that the Russians have accepted with respect to the use of airplanes.

The airplanes they use a lot are unpiloted cheap low-performance drones which they are willing to get shot down in huge numbers and they do get shot down in huge numbers. But piloted aircraft have played a limited role because the air defense environment is too lethal for an air force with skills no better than the Russians are to survive in it.

FASKIANOS: Thank you. I’m going to take the next question from Mike Nelson.

Q: Thanks for a very interesting overview. I work at the Carnegie Endowment for International Peace and also have taught at Georgetown on internet policy and the impacts of digital technologies.

Seems to me that one of the big changes with this war has been the incredible transparency, more information on what’s actually going on on the ground from social media, satellite photos, drone photos. I saw a tweet today about how they’re able to infer how many Russian soldiers have mutinied by counting these soldiers marching back from the front, presumably under armed guard.

It just seems that there’s a lot more information on what’s going on hour by hour. I wonder if that is causing some changes on both the Russian and the Ukrainian side and whether the insertion of disinformation to make it appear that things are going differently than it seems is also something that’s getting better and better.

Thank you.

BIDDLE: Yeah. I mean, the information environment in Ukraine is complicated in ways that the debate often doesn’t deal with very well, in my view.

So starting at the superficial level, public perceptions of what the lethality of first-person view kamikaze drones has been against tanks and artillery are wildly exaggerated and the reason why the public impression is wildly exaggerated is because the medium formerly known as Twitter puts up endless videos of successful attacks. But nobody posts a video of their failed attack so we only see the subset of all drone missions that succeeded.

The ones that don’t are invisible. Therefore, the public gets this impression that all—that there are successful drone missions by the millions all the time and that that’s—there are serious selection effects with the way the public understands drone success rates in light of that.

So one point is that the apparent transparency is subject to a variety of selection biases that lead to misunderstandings of the transparency on the battlefield as a whole. Similarly, there are lots of videos of images of Russian soldiers in a trench and especially videos of Russian soldiers in a trench before a quadcopter drone drops a grenade on them and then kills them.

You don’t see any video feeds of a drone flying over a camouflaged position where you can’t see anything because nobody’s going to post that, right? It’s not interesting enough. But, therefore, again, we get the selection effect. People believe that everything is visible and everything is transparent because every video feed they see, and they see a lot of them, shows a visible target.

The trouble is you’re not seeing the failed drone missions that didn’t produce a visible target and those are the vast majority as far as we can tell from more careful analyses that try to look at the totality of drone missions rather than just the selected subset that appear on now X, formerly Twitter.

Now, that leads to the general issue of how transparent is the modern battlefield and I would argue that the modern battlefield is a lot less transparent than people popularly imagine that it is. The cover and concealment available in the earth’s surface to a military that’s capable of exploiting it is still sufficient to keep a sizeable fraction of both militaries’ targets invisible to the other side most of the time and that’s why the artillery casualty rate hasn’t gone up dramatically as a result of all this. It’s because cover and concealment is still keeping most of the targets out of the way.

So I would argue the battlefield is less transparent than we often assume that it is and in part that’s because the systems that would generate information are countered by the other side so that they generate less information. Again, take drones, which have been the thing that everybody’s been focusing on. There have been multiple waves of measure/countermeasure races just on the technical side, setting aside technical adaptation, with respect to drones already.

When the war opened the primary drone in use, especially on the Ukrainian side, was the Bayraktar TB2, Turkish-built large, you know, capable, fairly expensive drone which was very lethal against exposed Russian armored columns.

Then several things happened. One is the armored columns decided to get less exposed. Smart move on the Russians’ part. The other thing is the air defense system under the Russians adapted and started shooting down Bayraktar TB2s at a huge rate to the point where the Ukrainians stopped flying them because they were so vulnerable and, instead, drones shifted from big expensive higher performance drones to smaller, cheaper, lower performance drones, which were so cheap that it didn’t make sense to fire expensive guided anti-aircraft missiles at them anymore and then the air defense environment shifted to emphasize jamming, which is even cheaper than the drones, and anti-aircraft artillery firing bullets that are cheaper than drones.

So the systems that would create this transparency and that would give you this information don’t get a free ride. The opponent systematically attacks them and systematically changes the behavior of the target so that the surviving seekers have less to find, and in addition to cover and concealment and complementary to it is dispersion and what dispersion of ground targets does is even if you find a target it may very well not be worth the expenditure of an expensive precision munition to kill.

A guided 155-millimeter artillery shell costs on the order of a hundred thousand dollars a shell. If you’re shooting it at a concentrated platoon of enemy infantry that’s a good expenditure. If you’re shooting it at a dispersed target where they’re in one- or two-soldier foxholes now even if you know where all the foxholes are—even if your drones have survived, the concealment has failed and the drone has accurately located where every single two-soldier foxhole is does it make sense to fire a $100,000 guided artillery shell at each of them or are you going to run out of guided artillery shells before they run out of foxholes, right?

So the net of all of this—the technical measure/countermeasure race and the tactical adaptation is that I would argue that the battlefield is actually not as transparent as people commonly assume. If it were we’d be seeing much higher casualty rates than what we’re actually seeing.

There’s incremental change, right? The battlefield is more transparent now, heaven knows, than it was in 1943. But the magnitude of the difference and the presence of technical measures and countermeasures is incremental rather than transformational and that’s a large part of the reason why the change in results has been incremental rather than transformational.

FASKIANOS: So we have a lot of questions but I do want to just ask you, Steve, to comment on Elon Musk’s—you know, he shut down his Starlink satellite communications so that the Ukrainians could not do their assault on the—on Russia. I think it was the submersible—they were going to strike the Russian naval vessels off of Crimea.

So that, obviously—the technology did affect how the war was—the battlefield.

BIDDLE: It did, but you’ll notice that Crimea has been attacked multiple times since then and metropolitan Russia has been attacked multiple times since then. So there are technical workarounds. On the technical side rather than the tactical side there are multiple ways to skin a cat. One of these has been that the U.S. has tried to make Ukraine less dependent on private satellite communication networks by providing alternatives that are less subject to the whims of a single billionaire.

But tactical communications of the kind that Starlink has enabled the Ukrainians are very useful, right? No doubt about it, and that’s why the U.S. government is working so hard to provide alternatives to commercial Starlink access.

But even there, even if you didn’t have them at all the Ukrainian military wouldn’t collapse. I mean, in fact, most military formations were taught how to function in a communications-constrained environment because of the danger that modern militaries will jam their available communication systems or destroy communication nodes or attack the satellites that are providing the relays.

Certainly, the U.S. military today is not prepared to assume that satellite communications are always going to be available. We train our soldiers how to operate in an environment in which those systems are denied you because they might be.

So, again, I mean, tactical adaptation doesn’t eliminate the effects of technological change—having Starlink, being denied Starlink, right, this Musk-owned communication satellite constellation that was the source of all the kerfuffle.

It’s not irrelevant whether you have it or not but it’s less decisive than you might imagine if you didn’t take into account the way that militaries adapt to the concern that they might be denied them or that the enemy might have them and they might not, which are serious concerns.

Certainly, if the U.S. and Russia were true belligerents both the danger of anti-satellite warfare destroying significant fractions of those constellations is serious, or jamming or otherwise making them unavailable is a serious problem so militaries try to adapt to deal with it—with their absence if they have to.

FASKIANOS: Great. We have a question—a written question from Monica Byrne at—a student at Bard College: Can you share thoughts and strategy for Israel and Gaza, given the conditions in Gaza?

BIDDLE: Yeah. So shifting gears now from Ukraine to the Middle East, given Israel’s declared war aim, right—if Israel’s aim is to topple the Hamas regime and then hopefully replace it with something that’s another conversation.

But let’s for the moment just talk about the military dynamics of realizing their stated war aim of toppling the Hamas regime. That will certainly require a ground invasion that reoccupies at least temporarily the entirety of Gaza, right? Airstrikes aren’t going to accomplish that war aim. Special forces raids aren’t going to accomplish that war aim. The Hamas administrative apparatus is, A, too large and, B, too easily concealed, especially underground, for those kinds of techniques to be sufficient.

So if the Israelis really are going to topple Hamas a large-scale ground invasion is needed. That has obvious horrible implications for collateral damage and civilian fatalities in Gaza—urban warfare is infamously destructive of capital and of civilian human life—but also for military casualties to the Israelis.

Urban warfare is a radically advantageous military environment for defenders and so Israel inevitably will take serious losses if they really expect to completely reoccupy Gaza as would be needed to depose Hamas.

Now, there are ways that conventional militaries can try and reduce either the loss of innocent civilian life or casualty rates to their own forces but none of these things are perfect and the techniques militaries use to reduce civilian fatalities can be exploited by defenders who want to take advantage of them to increase Israeli military casualties and limit the Israelis’ ability to limit collateral damage.

You can fire only at identified targets and not at entire buildings. You can use small-caliber weapons rather than large-caliber artillery and missiles. You can warn the civilian occupants of a building either with leaflets or text messages or the Israeli technique that’s called knocking on the roof where they drop a nonexplosive weapon on the ceiling to create a sound that tells the occupants they are about to be attacked so they leave.

There are a variety of things like that that you can do and that the U.S. should hope that the Israelis are going to do. But the whole problem here is that the Hamas political and military infrastructure is deeply intermingled with the civilian population in Gaza, and so even if you’re going to be as discriminating as modern technology and military skill potentially could make you, you’re still going to kill a lot of civilians and Hamas is not going to conveniently remove the military infrastructure from the civilian population to make it easier for the Israelis to kill the fighters and not kill the civilians. They’re going to keep them tightly intermingled.

Now, the Israelis can reduce their losses by being slower and more deliberate and methodical in the way they enter Gaza. There’s been a discussion in recent weeks about the difference between Mosul and Fallujah and the U.S. experience of urban warfare in Iraq.

In Fallujah, we entered quickly with a large ground force that was fairly dependent on small arms direct fire and relatively less reliant on artillery and airstrikes. In Mosul with Iraqi allies on the ground, we did the opposite. Very slow entry. The campaign took months. Limited exposure, small-caliber weapons, heavy emphasis on airstrikes and artillery to reduce the ground—even so, thousands of civilians were killed in Mosul. Even so, our Iraqi allies took serious casualties. There’s no way for the Israelis to do this Gaza offensive if they’re going to realize their war aim that won’t destroy Gaza, kill a lot of civilians, and suffer a lot of casualties themselves. All these things are marginal differences at the most.

FASKIANOS: Thank you. I’m going to go next to Dan Caldwell.

Q: Oh, Steve, thanks very much for a very interesting overview.

I’d like to raise another subject that is, obviously, very broad but I would really appreciate your comments on it and that’s the question of intelligence and its relationship to military operations that you’ve described.

Broadly speaking, we can separate out tactical intelligence from strategic intelligence, and in the case of tactical intelligence the use of breaking down terrorists’ cell phones’ records and things like contributed to military successes in Iraq and Afghanistan.

In a strategic sense, the breaking of the Japanese codes, Purple, and the Ultra Enigma secret in World War II contributed to the Allies’ success, and in terms of the Middle East the strategic failures of Israeli intelligence in 1973 and, I would argue, in the recent Hamas attacks contributed to the losses that Israel has suffered.

So how do you think about the relationship of intelligence to military strategy?

BIDDLE: Yeah. I mean, intelligence is central to everything in security policy, right? It’s central to forcible diplomacy. It’s central to preparation for war. It’s central to the conduct of military. So intelligence underlies everything. All good decision making requires information about the other side. The intelligence system has to provide that.

The ability of the intelligence system to create transformational change is limited. Let’s take the national level strategic intelligence question first and then we’ll move to things like Ultra and battlefield uses.

As you know, the problem of military surprise has been extensively studied, at least since the 1973 war in which Israel was famously surprised by the Egyptian attack in the Sinai. There’s been an extensive scholarly focus on this problem of intelligence failure and surprise—how can this possibly happen.

And the central thrust of that literature, I would argue, has been that almost always after a surprise you discover later that the surprised intelligence system had information that should have told them an attack was coming.

They almost always receive indicators. They almost always get photographic intelligence. All sorts of pieces of information find their way into the owning intelligence system. And yet, they got surprised anyway. How could this happen?

And the answer is that the information has to be processed by human organizations, and the organizational challenges and the cognitive biases that individuals have when they’re dealing with this information combine in such a way to frequently cause indicators not to be understood and used and exploited to avoid surprise and part of the reason for that—the details, of course, are extensive and complex.

But part of the reason for that is you get indicators of an attack that didn’t—that then didn’t happen way more often than you get the indicators of the attack that does happen. You get indicators all the time but usually there’s no attack and the trick then is how do you distinguish the indicator that isn’t going to become an attack from the indicator that is going to become the attack when you’ve always got both.

And if you—especially in a country like Israel where mobilizing the reserves has huge economic consequences, if you mobilize the reserves every time you get indicators of an attack you exhaust the country and the country stops responding to the indicators anymore. It’s the cry wolf problem.

I mean, the first couple of times you cry wolf people take it seriously. The eighth, ninth, tenth, twelfth time they don’t. So because of this the ability to change, to do away with surprise, with, for example, new technology, all right, a more transparent world in which we have a better ability to tap people’s cell phones and tap undersea cables to find out what governments are saying to themselves we have better ability to collect information.

But there are still organizational biases, cognitive problems, and just the basic signal-to-noise, wheat-to-chaff ratio issue of lots and lots of information, most of which is about an attack that isn’t going to happen. And distinguishing that from the ones that are going to happen is an ongoing problem that I doubt is going to be solved because it isn’t a technological issue. It resides in the structure of human organizations and the way the human mind operates to filter out extraneous and focus on important sensory information, and human cognitive processes aren’t changing radically and human organizations aren’t either.

So at the strategic level I don’t see transformation coming soon. Then we’ve got the battlefield problem of what about intercepted communications, for example, which have changed the historiography of World War II in an important way.

We’ll note that that didn’t cause the Allies to defeat the Germans in 1944, right? I mean, the Allies cracked the German and the Japanese codes long before the war ended and, yet, the war continued, and this gets back to this question of how militaries adapt to the availability of information about them on the other side.

At sea where there’s not a lot of terrain for cover and concealment, right, then these kinds of communications intercepts were more important and as a result the Japanese navy was, largely, swept from the Pacific long before the war ended in 1945.

But wars are ultimately usually about what goes on on land, and on land even if you intercept people’s communications if they’re covered, concealed, dispersed, and in depth being able to read German communications, which we could do in 1944, didn’t enable us to quickly break through, rapidly drive to Berlin and end the war three months after the Normandy invasions. In spite of the fact that we could read the communications traffic we couldn’t do those things because the communications traffic is only part of success and failure on the battlefield.

So if that was the case in World War II where we had, you know, unusually good comment and usually good ability to break the enemy’s codes and read their message traffic, again, I would argue that improvements in intelligence technology today were certainly helpful, and they’re worth having and we should pursue them and use them, but it’s not likely to transform combat outcomes in a theater of war any more than—to a radically greater degree than it did when we had that kind of information in 1944.

FASKIANOS: So I’m going to combine the next two questions because they’re about innovation from the Marine Corps University and Rutgers University: You mentioned over innovation. Can you explain what that is and how it can be detrimental? And then are you concerned that the Department of Defense R&D program could be at risk of being out of balance by over emphasizing advanced technology versus getting useful technology deployed and into the field?

BIDDLE: I think that’s one of the most important implications of this war is that the United States has historically chosen to get way out on the envelope of what technology makes possible for weapon acquisition, creating extremely expensive weapons that we can buy in very small numbers that we evaluate and we decide to buy because of their proving ground potential because what they can do against targets that haven’t adapted to them yet.

What the record of adaptation in Ukraine, I think, shows is that the actual lethality of very sophisticated weapons is not as high as it looks on a proving ground because the targets are going to be noncooperative and the real-world performance of extremely expensive sophisticated technologies is normally less than it looks, and if that’s the case we are probably overspending on very sophisticated, very expensive weapons which we can only buy in very small numbers and which if they don’t produce this radical lethality wouldn’t be worth the expenditure that they cost.

And if the adaptation of the target is going to reduce their lethality and increase their vulnerability, which is certainly what we’re observing in Ukraine, then we’re going to have a dickens of a time replacing them when they get lost, right, because very sophisticated high technology weapons, among other things, require a supply chain of materials that are often quite scarce—rare earths, cobalt, lithium.

One of the reasons why the American Defense Industrial Base has had a hard time responding rapidly to the demands that the expenditure rate of things in Ukraine has created is because of these complicated supply chains that we can manage when we’re building things in small numbers, which we think is sufficient because we’re expecting that each one of them is going to be tremendously lethal.

If we now realize that they’re less lethal in practice than we expect them to be and therefore we need larger numbers of them, how are we going to get the materials we need to do that? And the experience in Ukraine has been that the kind of revolution in military affairs expectation for the lethality of high technology just hasn’t been realized.

Yes, weapons are very lethal in Ukraine, but not orders of magnitude differently than they were in 1944, right, and so I think this ought to suggest to us that the historical post-World War II U.S. strategy emphasizing very high technology at very high cost in very small numbers to compensate for small numbers with radical lethality may very well be misguided.

It works well when you’re fighting an opponent like the Iraqis who can’t handle the complexity of cover and concealment, combined arms, and all the rest. They’re exposed and the weapons have the kind of proving ground effect that you expect because the targets are not undercover. Not clear that it has been producing that kind of results in Ukraine and it’s not clear that it would produce those kinds of results for the United States in a coming great power conflict.

FASKIANOS: Thank you. I’m going take the next question from Genevieve Connell at the Fordham graduate program in international political economy and development. How much does successful military strategy rely on stable domestic economic systems to fund it or is this less of an issue when one or both sides have strong geopolitical support and aid?

BIDDLE: War is very expensive, as the Ukraine war is reminding us, right? This isn’t news. The expenditure rates in modern industrial age warfare are massively expensive to maintain and that in turn means that the strength of the national economy is a fundamental foundational requirement for success in modern great power warfare.

This, of course, leads to the set of tradeoffs that are fundamental in grand strategy, right? Grand strategy, as opposed to operational art, military strategy, or tactics, integrates military and nonmilitary means in pursuit of the ultimate security objectives of the state and one of the more important of the nonmilitary means is the economy.

So you need a large GDP to support a large expensive war effort. The way you maximize GDP is with international trade. International trade makes you vulnerable to cutoff in time of war through blockade. Therefore, if we just maximize GDP in the short run we run the risk—we increase our vulnerability in time of war or blockades. We say: Oh, no, we don’t want to do that. Let’s reduce the amount of international trade we do, make ourselves more self-sufficient. Now GDP growth rates go down and now the size of the military you can support in steady state goes down. There’s a fundamental tradeoff involving the interaction between classically guns and butter in the way you design the economy in support of the grand strategy you have in mind for how you’re going to pursue your security interest in the international system at any given time.

So, yeah, a productive expanding economy is essential if you plan to be able to afford the cost of modern warfare. The implications for what that means for things like international trade, though, are complicated.

FASKIANOS: Great. I’ll try to sneak in one last question from David Nachman.

Q: Thank you. Thank you for this really interesting presentation. I teach at the Yale Law School, nothing related to the topic of today’s submission and discussion.

I’m just wondering, and you captured it towards the end here where you said something about wars are won and lost on land. With the advent of cyber and all the technological development that we’re seeing in our armed forces is that still true as a matter, you know, and are we—is the Ukraine and even Gaza experience sort of nonrepresentative of the true strategic threats that the United States as opposed to its allies really faces at sea and in the air?

BIDDLE: Yeah. Let me briefly address cyber but then extend it into the sea and the air.

One of the interesting features of cyber is it’s mostly been a dog that hasn’t barked, at least it hasn’t barked very loudly. There were widespread expectations as Russia was invading that cyberattacks would shut down the Ukrainian economy, would shut down the Ukrainian military effort, or vice versa, and neither of those things have happened.

So I don’t—there have been plenty of cyberattacks, right, and there have been plenty of efforts at break in and surveillance and manipulation. So far none of them have been militarily decisive and it’s an interesting and I think still open question for the cyber community about why that has been so and what, if anything, does that tell us about the future of cyber threats to national military projects. But so far it hasn’t radically—it hasn’t produced a result that would have been different in the pre-cyber era.

Now, when I say wars are won on land what I mean by that is that people live on the land, right? People don’t live in the air and people don’t live on the surface of the water. People live on land. Economies are on land. Populations are on land.

That means that usually the stakes that people fight wars over are things having to do with the land. That doesn’t mean that navies and air forces are irrelevant. We own a large one. I’m in favor of owning a large one. The Navy—my friends in the Navy would be very upset if I said otherwise.

But the purpose of the Navy is to affect people who live on the land, right? In classic Mahanian naval strategy the purpose of the Navy is destroy the opposing fleet, blockade the enemy’s ports, destroy the enemy’s commerce, and ruin the land-based economy and it’s the effect of the land-based economy that causes surrender or compromise or concession to the opponent or whatever else ends the war in ways that you hope are favorable to you.

What this means then is that especially where we’re dealing with large continental powers like Russia, classically—China’s an interesting sub case but let’s talk about Russia—the ability to influence the Russian decision-making calculus that leads to an end to a war or the beginning of a war without affecting the life of people on land is very limited.

Cyber has not proven able to do that. Air attack historically has not been a good tool for doing that. Navies do that by affecting the land-based economy and I don’t see that changing rapidly anytime soon.

FASKIANOS: Well, Steve, thank you very much for this really insightful hour. I’m sorry to all of you we couldn’t get to the questions, raised hands, so we’ll just have to have you back. And thanks to all those of you who did ask questions. I commend to you, again, Steve Biddle’s Foreign Affairs piece, “Back in the Trenches,” and hope you will read that.

Our next Academic Webinar will be on Wednesday, November 8, at 1:00 p.m. (EST) with José Miguel Vivanco, who is an adjunct senior fellow here for human rights, to talk about human rights in Latin America.

So, Steve, thank you again.

BIDDLE: Thanks for having me.

FASKIANOS: And I—yes. And I’d just encourage you all to learn about CFR paid internships for students and fellowships for professors at CFR.org/careers. Our tenured professor and our fellowship deadlines is at the end of October. I believe it’s October 31, so there’s still time. And you can follow us on X at CFR_Academic. Visit CFR.org, ForeignAffairs.com, and ThinkGlobalHealth.org for research and analysis on global issues.

Thank you all again for being with us today.

(END)

Top Stories on CFR

Energy and Climate Policy

What powers artificial intelligence (AI)? As global electricity use is surging, with unprecedented demand coming from an increase in data centers, AI’s dependence on fossil fuels presents a serious issue for the planet. In the United States, data center power usage is on track to double by 2030, largely due to the proliferation of AI technology. But while the application of AI shows potential to mitigate climate problems through modeling or predicting weather events, will its power grab stall the clean energy transition?

South Korea

Yoon Suk-Yeol’s shocking declaration of ‘emergency martial law’ was a frontal assault on the integrity of South Korea’s hard-won democratic system.

Lebanon

An array of domestic and foreign powers are vying for influence in Lebanon, including the Lebanese Armed Forces, Hezbollah, Israel, Iran, Syria, and the United States.