Report 055 · Defense Tech
Counter-IED jammers are one of the clearest engineering successes of the Iraq and Afghanistan wars. They did the job they were built for. The bombs kept killing people anyway, and the reason is not a failure of the jammer. It is a lesson about what a countermeasure is, which almost nobody states plainly.
This was my job. Counter-IED and electronic warfare, which in practice means thinking very hard about a radio link that somebody else does not want you thinking about. So when I read that some new system "defeats" a threat, I have a specific reflex, and it comes from watching the most successful counter-IED technology of the era work exactly as designed and not end the problem.
I want to walk through why, because the shape of it repeats everywhere. It is the same shape as the drone story I have written about more than once, and once you see it you cannot unsee it in a product brochure.
Start with the mechanism, because the mechanism is the whole argument.
A radio-controlled improvised explosive device has a receiver in it. A phone, a car alarm fob, a doorbell chime, a toy controller. Someone watching the road presses a button, a signal goes out, the receiver hears it, and a circuit closes. The bomb is a receiver with explosives attached.
The counter is to make the receiver deaf. Vehicle-mounted jammers, the family that became the CREW program, radiate energy across the bands those receivers listen on so that the trigger signal never resolves into a clean command. The Congressional Research Service described the early systems, Warlock and ICE, in plain terms: they "use low-power radio frequency energy to block the signals of radio controlled explosives initiators, such as cell phones, satellite phones, and long range cordless telephones."
That works. Not in a hedged, marketing sense. Physically, reliably, on the correct frequencies, it works, and it is one of the fastest fielded engineering responses in modern military history. The same CRS report also names the constraint that matters, and it is worth reading slowly, because it is the seed of everything that followed: experts "caution that the jammers may only be partially effective because they must be set to operate within the right frequency range in order to stop an IED."
A jammer is not a shield. It is a denial of one specific channel. Every watt it spends is spent on the assumption that a radio link exists to deny.
The urgency behind that assumption is easy to lose two decades later. The Government Accountability Office laid out the curve in its 2009 review of the Defense Department's counter-IED effort: "Beginning in June 2003, IED incidents targeting coalition forces began to escalate from 22 per month to over 600 per month in June 2004. In June 2006, these incidents reached more than 2,000 per month." At the peak, GAO notes, "coalition forces in Iraq were experiencing almost 100 IEDs per day." By 2008, IEDs accounted for almost 40 percent of the attacks on coalition forces in Iraq.
Against numbers like that, a countermeasure that reliably kills one initiation method is not a small thing. It is a very large thing. It just is not the thing people assume it is.
Open-source ordnance researchers categorize IED initiation the same way military doctrine does, in three families. Action on Armed Violence, in a review written with the explosives specialist Roger Davies, sets them out cleanly. A timed device "is initiated by some form of timing mechanism, set by the perpetrator, so that they can remove themselves from the scene." A command-initiated device fires "at the moment the terrorist decides," and the initiation "might be by command wire, by radio control or as a suicide switch." A victim-operated device is "set by the perpetrator so that an unknowing act by the victim causes the device to function," through something like a trip wire or a pressure pad.
Now count what the jammer reaches. Not timed devices, which need no signal at all. Not victim-operated devices, where the person who dies closes the circuit with their own boot or their own front tire. Not even all command-initiated devices, because a command wire is a piece of copper running back to a person in a ditch, and there is no radio-frequency energy on earth that will interfere with a piece of copper.
One branch, of one family, out of three. That is the honest coverage map of a technology that genuinely works.
What happened next is the part I think is genuinely instructive, and it is documented rather than folklore. The AOAV review states the dynamic directly:
"Ever since 'jammers' were deployed by those defending against such attacks, terrorists have explored ways of getting around such defenses, either by moving the frequency of the command system to an area of spectrum outside the jammer's capability or using a combination of switches."
Two distinct escapes in one sentence, and they are worth separating. The first is a spectrum race: move the trigger to a band the jammer is not covering, then the jammer chases, then you move again. That race is winnable in the short term by whoever can reprogram faster, and it is the reason CREW systems went through generation after generation.
The second escape does not race at all. It leaves the spectrum entirely. Reporting from Afghanistan in 2010 described exactly that transition: as coalition forces mounted radio and cell phone jammers on vehicles to stop remote detonation, insurgents moved to a wire connecting the bomb to the detonator, the command wire. And alongside it, victim-operated pressure plates, against vehicles and against troops on foot. When metal detectors began finding plates built from large metal contacts, the builders switched to smaller metal filaments on wood or plastic, cutting the metal signature the detector was looking for.
Look at what that sequence is. Each countermeasure is answered not by beating it, but by deleting the thing it detects. The jammer looks for a radio link, so remove the radio link. The detector looks for metal, so remove most of the metal. You do not defeat the sensor. You stop giving it anything to find.
The analysts at CNA, who supported counter-IED work in both theaters from 2004 onward, summarized the whole decade in one line: "Again and again, insurgents adapted to counter-IED tactics, improving their technology. U.S. forces answered back with new counter-tactics, each demanding analysis to learn which were most effective."
Here is where I part company with the cynical reading, because "the jammers didn't work" is as wrong as "the jammers solved it."
Forcing an adversary off the radio is not nothing. It costs them real capability, and the cost is worth naming precisely.
A radio-controlled device lets the attacker choose the moment. They can sit off at a distance, watch a convoy, wait for the vehicle they want, and fire. That is a discriminating weapon operated from relative safety. Take the radio away and the alternatives all impose a penalty. Command wire puts a human being physically close, connected to the device by a line that can be seen, traced, and followed back to them, which turns the trigger into a targeting opportunity. A pressure plate removes the operator's risk but also removes their judgment: the device now fires for whoever steps on it, which means the local civilian on the same road, on their own schedule, with no one watching.
That last consequence is not an abstraction, and it should not be laundered into a tidy tactical trade. Pushing an adversary toward victim-operated initiation makes the weapon indiscriminate, and the people who live on that road inherit the difference. A countermeasure can succeed on its own terms and still displace harm rather than remove it. Both of those things are true at once and any account that keeps only one of them is selling something.
So the correct scorecard for a countermeasure is never "did the attacks stop." It is: which capability did we take away, what did the adversary have to accept in order to keep operating, and are we now positioned against the thing they moved to. Those questions have answers. "Did we defeat the threat" does not.
A few things I am not claiming. I have not put a percentage on the trigger-type shift, because the credible open sources describe the direction of the change rather than a clean before-and-after breakdown, and the tidy vendor statistics that do circulate come from companies selling the systems. Direction is well supported. Precise proportions, in open sources, are not, and I would rather say so than borrow a number I cannot stand behind.
The GAO figures above are incident counts for coalition forces in Iraq, not a global measure and not a casualty count. The AOAV and CNA material is analytical rather than statistical. And the Afghanistan reporting on command wire is contemporary journalism from 2010, which I treat as a well-sourced account of what was observed at the time rather than a doctrinal statement.
A countermeasure is a bet about a mechanism. The jammer is a bet that the trigger uses radio. The metal detector is a bet that the switch contains metal. The bet can be entirely correct and still stop paying, the moment the adversary is willing to give up the capability that made the bet true.
That is the question to ask of any defensive system, and you can ask it without any classified knowledge at all. Not "does it work," which is usually a solved engineering question and usually yes. Ask what physical assumption it rests on, and what it costs the other side to walk away from that assumption. If walking away is cheap, you have bought yourself a spectrum race. If walking away costs them precision, or safety, or reach, you have bought something real, and you should be able to name exactly what it is.
The jammers bought something real. It was just never the thing the word "defeat" implies.
U.S. Army combat veteran (Counter-IED / Electronic Warfare), peer-reviewed researcher in microwave spectroscopy, and founder & CEO of Shroombiosis. Consults on laboratory operations, AI, and supplement formulation.