On February 16, 2001, U.S. military aircraft attacked several sites inside Iraq. Their stated goal was to disrupt Iraq's ability to target Coalition aircraft enforcing the no-fly zones with ground-based weapons. In the days following the attack, it came to light that the U.S. Navy had used the AGM-154 JSOW to attack approximately twenty-eight of the targets selected. Some days after that, it became known that (apparently) many of the JSOWs missed their aimpoints, and only around half of the targets attacked were hit.

This provoked a variety of responses from analysts around the world, but most notably in the United States press. One such example was published in the "Dot.Mil" column in the Washington Post, written by columnist and analyst William Arkin. This column made several technical and normative arguments which I found confusing; I therefore wrote Mr. Arkin for his response to my questions. Although he hasn't responded yet (not enough time) if and when he does, I will add his responses here. UPDATE: Mr. Arkin has given his permission to post his responses to my questions! (If you feel the need to verify it, his email is at the bottom). Thanks Mr. Arkin...

In the meantime, first read his column at:


...if the link above ceases to work, I will ask his permission to post the column here for purposes of argument.

...done? Okay, good. Here's the email letter I wrote to him on 2/26/01 late in the evening, and his responses. I’ll put his responses in indented italics and preface them with WA:.

Mr. Arkin-

I was most interested to read your column on the JSOW use in the recent Iraq strikes. I have a couple of questions that I'd like to pose to you regarding your article, as well as a few general observations (of course, ignore me at will! :-))

First of all, the fact that the JSOW is a cluster-only weapon is correct...but only technically. The JSOW as ordered has three variants, the AGM-154A, -B and -C variants. the A variant is the antipersonnel version. The B version, while also containing submunitions, uses BLU-108 Anti-armor/Anti-personnel munitions. While this would be a more appropriate choice, it is not due to be available (according to US Navy PR) until FY2002. The third, type C, is a unitary warhead which is ideal for point and hardened sites like air defense- but is not expected to reach IOC until 2003 (data from http://www.fas.org, http://www.chinfo.navy.mil/navpalib/policy/vision/vis99/v99-ch3c.html) From your article it would seem that the navy chose to drop type B's (you mention anti-armor/anti-personnel bomblets) which implies that in fact they were likely to have been the best JSOW option available.

    WA:>The BLU-97 sub-munition in the A version is officially an anti-armor/anti-personnel bomblet. The B version carries sensor fuzed weapon payloads, which are more akin to mines. The A version is the one, in a strict military analysis, that would be the best choice to hit radar and SAM sites to cause a higher level of damage than with HARMs.

The question then becomes, given the choices available, why did the Navy choose the JSOW at all? Is it possible that the Navy (which hasn't officially commented) was aware of potential problems with winds, and chose the area attack weapon to try to offset this?

    WA:Good question. The workup of the target obviously didn't take into account the winds or the rotation of the earth for that matter (all missiles landed "left," because they were shot south to north). JSOW has been used, fairly successfully, since Jan 1999. What is interesting though is that without a gun camera (or report back), the only real time information about its "success" is performance data that the missile left the airplane and went to the target, and imagery after the strike that shows the impact of cluster bombs, which I imagine is fairly elliptical and subjective. Is the Navy "experimenting" with the weapon in Iraq? Absolutely, experimenting two fold, with the weapon, and with the new "DEAD" (as opposed to "SEAD") strategy.

Was there an availability issue with the JDAMs? The Navy notes that the AF is the lead service for JDAMs. the JDAM also does not offer the standoff range of the JSOW-

    WA:JDAM does not have as great a stand-off range, but it is still siginficant. And it is not the only unitary stand-off weapon available. I believe the choice was to use a cluster bomb, not use JSOW as opposed to another stand-off weapon.

-it would be instructive to know the estimated air defense threat around the JSOW targets vs. the JDAM and other weapon targets. Do you have any information on that?

    WA:I know that there is some concern that AAA can shoot down a JSOW because of the dive angle.

Later on in your article, you comment that the JSOW being used as a weapon of choice for air defense counterstrikes seems to indicate that "...years of bombing in Iraq have had less than spectacular results on Iraq’s air defenses and the U.S. military is looking for some way of causing more permanent damage to the country's military capabilities." My question here is why is this an issue?

The issue really should then be why they didn't use these type of things earlier, and why (as you mention) the previous bombing was ineffective. If it was, then it seems to be a strong motivation to try different solutions, especially ones which have a better chance of causing 'more permanent damage.'

The reasoning that the JSOW might be better suited to destruction than suppression is not only based on the fact that it uses bomblets, but on the fact of its silent approach and long standoff range. the problem with antiradiation missiles is that they only really work if the radars in question are on! To quote the FAS again, from http://www.fas.org/man/dod-101/sys/smart/agm-88.htm:

"The primary lethal Supression of Enemy Air Defense (SEAD) platform, the F-16 employing AGM-88 High Speed Anti-radiation Missiles (HARM) has several shortfalls. It is becoming increasingly difficult to logistically support the F-16 and the HARM. SEAD forces have limited automated mission planning capability. It is very difficult to stimulate, decoy, and saturate enemy threat radars without putting friendly forces in harm's way, and the ability to reactively target surface-to-air threats is limited. The ability to employ off-board targeting sources is limited in the timeliness and accuracy required for the preemptive destruction mission. Though offboard sources may find mobile targets, there is a limited capability to pass required information in real time so fighters can reactively or preemptively target mobile surface-to-air threats. There is no on-board capability to preemptively target mobile surface-to-air threat systems. Current SEAD weapons all depend on RF homing for guidance and are vulnerable to emission control (EMCON) countertactics."
I note that the F/A-18 is also a HARM-shooter, so obviously platforms aren't the reason the Navy chose not to use them. The use of HARMs and other RF-seeking weapons to destroy or degrade (as opposed to suppress) enemy air defense installations is limited by the fact that they are a short time-window weapon. If you are just intending to suppress the defenses in order to get a strike package past them, then the HARM is quite useful; it will likely kill some of them, and at worst will likely force the enemy to use EMCON tactics, temporarily reducing his own effectiveness - which is the point. Permanent damage requires a different approach, and the use of low-warning, directed (as opposed to seeking) weapons seems to me a much better route. Also, if the target does not contain a working RF emitter (control points, command facilities, missile dumps, etc) then the HARM is completely useless. Were the sites struck all radars?

    WA:No. And even if HARM is not used, up to now, Mavericks have been the SEAD weapon of choice. It also has a small warhead.

I must admit I didn't check that fact carefully, and might have difficulty getting the data in any case.

Next, you comment that "The use of cluster bombs against minor, out of the way targets...actually helps Iraq to achieve its foreign policy goals." Does this imply that the unitary weapons were used on not-so-out of the way targets?

    WA:No, all weapons were. And I'm looking into whether the targets weren't selected because of their location as opposed to their importance.

If so, then perhaps the weapon choice was made with some thought to civilians in mind; better to use them on air defense sites in the boondocks than on those situated on urban rooftops, after all.

    WA:Sure, but I don't think that is the calculation that is made.

Finally, you take issue with the fact that the stated mission result was 'degraded' the Iraqi air defenses. The military here is hampered; they know they are using less accurate weapons than HARM; they know they are forced to use cluster versions rather than the unitary warheads they might prefer because they (maybe) aren't available. They are under a strong pressure to avoid casualties while achieving some result; they thus opt for lower probability of kill but cheaper (in dollars and risk) weapons, figuring that they can always revisit if need be. What is their option; go for 'destruction' when they (may) know they can't promise or achieve that? Claim destruction when they know they didn't hit?

    WA:There is no question that the military has improved its tactics, its ROEs, and its technology to pursue the impossible mission here. The result, beyond numbers of radars and SAM sites destroyed, is minor.

If your objections (as some of them appear to be) are to the policy decision that resulted in the Armed Forces being ordered to strike these air defense sites, as opposed to the choice of methods by the services, then I promptly withdraw all argument; that's an entirely different story. However, it appears that your objections seem to rest both on the policy and the execution, and it seems to me that given the constraints the military of the US is made to operate under, some of which you point out, then debating the choice of weapons is missing the point. That is, unless you are arguing that the policymakers are dictating specific weapon uses. If that's the case, again, I withdraw; I would characterize that as categorically stupid. We have a professional military; tell them what to do, and leave it to them to decide how to do it.

    WA:As you know, this is never the case. There are strict collateral damage constaints in the NFZ missions. The military can't be told to achieve a goal because there is no goal, and because this isn't a classic military operation even with an objective. Every day is a new mission, and the connection between yesterday's "success" and overall success is tentative.

In any case, thank you very much for your column; it is always good to see facts such as those you point out aired in public, especially on a topic with such strong ethical currents as the use of lethal force by our nation.

Oh, BTW, my background (in case you are wondering) is that of a military analyst, not a serviceman; I have worked for the RAND corporation and the Charles Stark Draper Laboratory doing military analysis on various forms of smart and competent munitions. I am a PhD student in International Relations and Strategic Studies. (remainder redacted)

Sincerely, 'The Custodian' (Yes, I used my real name when writing to him.)

    WA: I enjoyed answering your questions.

    Bill Arkin

Mr. Arkin’s response to my request for permission to post his replies:

I wrote it, and I appreciate your asking, so I guess the answer is yes.

Bill Arkin


Log in or registerto write something here or to contact authors.