The impact of house demolitions on Palestine suicide bombing

This paper examines whether house demolitions are an effective counterterrorism tactic against suicide terrorism.  We link original longitudinal micro-level data on houses demolished by the Israeli Defense Forces with data on the universe of suicide attacks against Israeli targets.  By exploiting spatial and time variation in house demolitions and suicide terror attacks during the second Palestinian uprising, we show that punitive house demolitions (those targeting Palestinian suicide terrorists and terror operatives) cause an immediate, significant decrease in the number of suicide attacks.

The effect dissipates over time and by geographic distance.  In contrast, we observe that precautionary house demolitions (demolitions justified by the location of the house but not related to the identity or any action of the house’s owner) cause a significant increase in the number of suicide terror attacks.  The results are consistent with the view that selective violence is an effective tool to combat terrorist groups, whereas indiscriminate violence backfires.

A new paper from Benmelech,  Berrebi, and Klor.

9 thoughts on “The impact of house demolitions on Palestine suicide bombing

  1. That’s very interesting!

    I can’t read the study in full (at least not for free), but I wonder if the authors considered the effect of the security barrier between Israel proper and the West Bank which is also creditted with having brought down the number of suicide attacks by more than 90 %.

    From what I have experienced in the Wes Bank, selective violence even faces less questioning from Palestinians than collective non-violent emasures (like the security barrier), although I would like to stress any country’s right to secure its borders in that respect.

  2. A nice video from ctrl-alt-shift about that barrier. I thought the London setting was very effective:

  3. Wonderful: A perfect example of scholarship used to examine the tree bark without looking at the forest. Or more cynically: how can we decrease the violence against us while continuing to occupy millions of people and colonize their land? Never mind that house demolitions are collective punishment and obviously illegal, not to mention immoral.

    Perhaps the next study should be on the effectiveness of torturing the children relatives of suicide bombers. If we couch it in academic quantitative language, maybe it’ll won’t sound like what it is.

  4. @ Sean: I want evidence to have a place in policy. Clearly the moral argument hasn’t carried the day when it comes to housing demolitions in the Israel/Palestine conflict. Perhaps an argument based on a demonstrated lack of efficacy might make sense. Of course this piece sounds soulless, it’s economic analysis.

    Doing the research on things like housing demolitions, torture and the like don’t mean there isn’t a moral argument to be made about these actions. It just means that there might also be a well supported argument about efficacy, or the lack thereof, to be made. Considering if something is even effective makes sense both from an academic point of view, but also from an activist point of view if one applies the lessons creatively.

  5. No, I’m afraid arguing against things like torture and house demolitions from the perspective of efficacy is a mug’s game, since the argument implies that if it’s effective, then it might be acceptable. Furthermore, as far as social science goes, it seems superficial in that it’s not looking at root causes of violence but rather immediate proximate factors.

    Frankly, articles like this, which seems to clearly bolster the Israeli government’s argument for illegal house demolitions, make me wonder if I should reconsider my opposition to academic boycotts.

  6. Chris,
    I think to post an abstract like this, without any qualifying comment is a dangerous game to play. Demolishing houses of civilians, regardless of their blood relations, is morally abhorrent, and there should be no mistake about it.
    Imagine someone posting the article: “Executing murderers’ immediate families along with the murderers themselves reduces crime rates in the US.” What kind of statement would putting up such a post without further commentary make?

  7. That indiscriminate attacks are ineffective is, in itself, an important point. On the other hand, what if the authors found that more discriminate attacks raised the risk of terrorism, and thus the policy prescription would be for fewer demolitions overall. Is this a publishable result? Does the implication of the result drive what should be published?

  8. In counter to that, right now we don’t know what is or isn’t effective, and yet governments (in this case Israel) are implementing these policies anyway. This research isn’t encouraging a government to try something both immoral and novel. It’s studying existing behavior.

    I’m just not sold on how having more information is a bad thing in a case like this. I honestly don’t see how it bolsters their case either. It undermines part of it pretty much entirely, and for selective demolition it indicates a rapid fade out of effectiveness. That doesn’t seem positive. I do get the concern that research can be twisted to evil purposes, but virtually any information can be twisted to evil purposes with enough motivation.

    To link in with comments from below, how do we decide what should or shouldn’t be published? I’d rather side with academic freedom to publish work. If we limit what we study that is bad, if we limit what we publish based on findings that seems even worse.

  9. Chris.

    I think it is pretty brave to post this abstract in the first place, this issue is not exactly uncontroversial…

    And it seems you sparked some interesting debate too, I look forward to follow the developments here. And I hope you have some input too.

    I don’t have access to the entire paper, but it would be very interesting to look deeper into which particular factors they have controlled for, and which factors they have not controlled for, in the study. The security barrier is already mentioned, but what about the growth of the non-violent movement in certain geographic areas, and the strengthening of Hamas and other religious movements in other geographic areas?

    How is it controlled to the time variable? Suicide bombings have indeed decreased massively over the past years.