Guest Post: A Rejoinder to Megan MacKenzie

by on 2012-10-19 in Duck- 7 Comments

Guest Post by Andrew Mack, Human Security Report Project.

We obviously agree with Megan Mackenzie’s affirmation that the mainstream narrative on wartime sexual violence ignores sexual violence against males and domestic sexual violence––even though the latter is far more prevalent than rape by combatants.

With respect to other claims in her review of our just-released Human Security Report we disagree:

We did not “revive” the “war is declining” thesis about inter-state conflict —the focus of successive Human Security Reports has been all conflicts—the overwhelming majority of which are intra-state.

Mackenzie claims that intra-state violence, political violence, and terrorism are increasing. Not so. The evidence from the UCDP/PRIO armed conflict dataset indicates that while minor armed conflicts have kicked up in recent years, the average number of ––mostly intrastate––high-intensity conflicts (those that result in 1,000 battle deaths per year) has halved since the late 1980s. Today the average conflict incurs less than a thousand reported battle deaths a year––a huge reduction from the Cold War years and far less than at the end of the 1990s.

The number of campaigns that seek intentionally to kill civilians was lower in 2009 than at any time since 1989, (which is as far back as UCDP has data). The number of victims from those attacks has also been trending downwards. NCTC data indicate that terrorist attacks were 30 percent fewer in 2011 than 2007.

Finally, the data on conflicts that don’t involve governments as one of the warring parties—those between rebel groups, militias, and other non-state actors—show neither a decline, nor the increase that Mackenzie suggests. All these trends (except terrorism) are reviewed in detail in Part II of the Report.

On the Report’s specific claims about wartime sexual violence:

Our “overarching message” is not that “wartime sexual violence is declining”. Rather, we argue, among other things, that there is no credible evidence to suggest that it is increasing, as many major reports and senior officials––all cited in the Report––have claimed. One of our conclusions is that the indirect evidence suggests that conflict-related sexual violence likely declines when the number and deadliness of conflicts declines. Mackenzie’s post conveniently misses these important qualifiers.

We do suggest that in general there is an association between conflict terminations and reduced levels of conflict-related sexual violence. This is hardly surprising. It’s clear, for example, that the level of conflict-related sexual violence (however defined) in Rwanda today isn’t remotely comparable to the mass savagery that was evident the height of the genocide/civil war. We do not of course argue that there is a linear relationship between battle death reductions and reduced levels of conflict-related sexual violence.

If it is indeed the case that conflict-related sexual violence tends to decline when wars stop, then seeking to stop wars may be an effective way to reduce such violence. There is after all little evidence to suggest that, in the short and medium term, any of the direct strategies currently being deployed to prevent, reduce and stop sexual violence are having other than a marginal impact. On the other hand there is considerable evidence that strategies to stop wars and prevent them from re-starting, have achieved a modest, but important, success rate. (See, Chapter 4 in “The Causes of Peace” in the last Human Security Report.)

We do argue––drawing on Dara Cohen’s important new global dataset––that the countries worst-affected by sexual violence constitute a minority of all war-affected countries.

In the new millennium, the level of reported conflict-related sexual violence reaches the highest of four intensity levels in less than 10 percent of the years in which countries experience conflict. In more than half the country-years of conflicts, the levels of reported sexual violence are low to minimal.

Only a handful of researchers have asked why some war–affected countries have very low levels of sexual violence. Yet understanding this could be extremely important for violence prevention policies––as Elisabeth Wood has pointed out.

On ‘strategic’ rape, our point is that, notwithstanding numerous claims by high-level officials and major reports (all cited in the Report) there is no evidence that it is increasing. We have no disagreement with Mackenzie on this issue. There is however some evidence from a remarkable multi-country study of sexual violence in sub-Saharan Africa by Dara Cohen and Ragnhild Nordas, which suggests that strategic rape may have declined in the region since the 1990s. We cite this research in the Report.

Finally Mackenzie writes that, “The HSR begins their report by chiding advocates and those who do research on sexual violence for drumming up unsubstantiated catchy headlines on sexual violence.” This is quite untrue. In fact the Report begins with a tribute to the extraordinary political campaign by advocates that got the issue of wartime sexual violence onto the agenda of the Security Council as a “threat to international peace and security.”

Moreover it should go without saying that, while a deep knowledge about sexual violence in war zones is essential to understanding individual cases, the findings from particular cases can neither prove or disprove claims about whether sexual violence in wartime is increasing or decreasing at the global level.

The 2012 Human Security Report draws on the best available data on sexual violence in wartime, although as we point out, the knowledge gaps remain huge. All the empirical claims in the Report are, however, subject to rigorous fact-checking.

We invite readers to judge for themselves whether Mackenzie’s criticisms are fair and correct. All chapters of the Report can be downloaded free of charge at

Print article

  • Lena Ohlsson

    Although I thought at first, with some of the headlines and with the critique on this blog, that this report could have a negative impact by enabling governments to reduce funding, etc, I now believe that’s not the case. Perhaps, as the majority of reviewers have stated, this will lead to more informed policy-making, and instead govts will be better equipped to help victims of sexual violence in conflict, especially those victims that often go overlooked. I also agree with Andrew Mack in that stating that conflict driven sexual violence hasn’t increased in recent years, like many advocacy agents claim, is not a bad thing if it’s accurate. It’s a good thing, and we should be working towards continuing to ensure that it declines worldwide. We can’t take the worst cases and apply those norms to every case.

  • Megan H Mackenzie

    I’m thrilled at the
    interest my post has received. There’s so much to say, but let’s stick to some
    of the main points.

    First, Dr. Mack you reiterate the war decline hypothesis but extend it. I’m surprised, in 2012, that you are using the decline in battle related deaths as an indicator of the incline or decline in war. So much evidence about the changing nature of war tells us that battle deaths are down, civilian deaths are up (and there are numerous counter studies to the one you cite- which take into account each year- indicating that civilian casualties overall are increasing).

    But let’s leave your war decline hypothesis for now.

    You reiterate that the overarching theme of the report was not that sexual violence is declining, but you do re-state that “conflict-related sexual violence likely declines when the number and deadliness of conflicts declines.” and
    “in general there is an association between conflict terminations and
    reduced levels of conflict-related sexual violence.”

    Be sure that I didn’t miss these qualifiers- I disagree with them and so do many other researchers (see WILPF and Oxfam’s response). You are assuming a fundamental link between the level of wartime violence and the level of sexual violence. You also assume that many significant conflicts are ‘exceptions’, including the Congo (with a report from the Guardian today that Congo rape numbers are increasing dramatically).

    There is evidence that the levels of conflict-related sexual violence are completely unrelated to other forms of violence AND do not necessarily decline with the end of war. For example, Dara Cohen, who the report does cite extensively, concluded “it is quite plausible that lethal violence is not closely correlated with sexual violence. In a pilot study of a sample of African conflicts—in research coauthored with colleagues at the Peace Research Institute Oslo (PRIO)—we find that named armed groups are still reported as perpetrators of sexual violence even after the lethal violence has stopped or nearly stopped. This suggests to me that killing and sexual violence are not necessarily correlated.” Research in post-conflict zones like Sierra Leone and Liberia has indicated that when the guns stop, sexual violence does not. For example in Liberia, almost a decade after the war ended, rape remains the number one reported crime with no signs of its decline. I guess it depends how you define conflict-related sexual violence. HSR want to draw attention to hidden forms of sexual violence like domestic violence and rape within the family- such a
    progressive move is countered by defining war-time sexual violence as something
    that only occurs while guns are blazing.

    As for the headlines counter-point (like others), we’ll have to just agree to
    disagree. Perhaps the intention was not to drum up headlines; however, the #1
    headline that has resulted from the report is “Wartime Sexual Violence
    Declines” just as the #1 headline from the 2005 report was “War
    Declines”. The truth is that whether sexual violence is indeed declining
    or not (which I clearly don’t believe it is), headlines like this are a
    distraction from researchers truly trying to understand the complexity and
    extent of this type of violence.

  • Megan H Mackenzie
  • ATRWibben

    Unfortunately you do not link to the reply by WILPF (which can be found here: and includes further replies to the report from several IR scholars who dispute its “findings”). Most strikingly WILPF points out that the HSReport actually states:

    “Although there are no reliable cross-national trend data on sexual violence in wartime, indirect evidence suggests that its incidence has declined worldwide over the past two decades. It is likely that conflict-related sexual violence decreased along with the decline in the number and deadliness of armed conflicts.”

    WOW. “Indirect evidence suggests” … “it is likely” – from a report that prides itself on using important new datasets and repeatedly points to the importance on reliable evidence comes the admission, right there, that its headline grabbing claim about a decline in sexualized violence is “likely,” based on “indirect evidence.” Enough said.

  • ATRWibben

    What is more, the Oxfam response which you also point to (and which can be found here: ends with a note on the possible policy impact of the HSR: “It would be tragic if its argument that ‘the level of sexual violence
    worldwide is likely declining’ encourages cash-strapped governments to
    stall the greater priority that sexual violence has, quite
    rightly, taken in the last few years.”

  • Pingback: The Cursory Pedant: War Rape, the Human Security Report and the Calculation of Violence « The Disorder Of Things

  • Pingback: Is Wartime Rape Declining On a Global Scale? We Don’t Know — And It Doesn’t Matter | Political Violence @ a Glance