THE 16 days campaign against violence against women and children has ended as it began, like TS Eliot’s world: not with a bang but a whimper.

There have been the usual indignant and politically correct speeches, but very little in the way of practical proposals to deal with a problem which research reveals to be pretty intractable.

In a paper published in August, five researchers from the Medical Research Council, the Western Cape Forensic Pathology Services and the University of Cape Town’s Forensic Medicine and Toxicology unit compared domestic violence in 1999 with 2009, based on mortuary statistics. They found that while in 1999 a woman died at the hands of her partner on average every six hours, by 2009 this number had dropped to: one every eight hours.

Progress? Technically yes, but it’s not much to shout about, especially since the researchers point out that though the number of women killed in homicides has dropped, the proportion of those murdered by intimate partners has increased (from just over 50 percent to 57 percent). And this is probably an under-estimate because in one in five murders the killer was never caught. The murder rate of women by their partners in South Africa is more than double that of the United States.

The researchers point out that if national efforts to prevent gender-based violence had been effective, there would have been a reduction in the proportion of women killed by their partners. So much for the campaigns and the speeches and the promises.

Fortunately, some useful steps have been taken. New “victim-friendly” rooms have been built in a few local police stations and more are to come. And commendable efforts have been made to train police to deal with complaints of domestic violence.

But much more has to be done and it is a pity that once again the the Minister for Women, Children and People with Disabilities, who leads the charge on this campaign, has missed an opportunity to implement a few big ideas.

More shelters for women and children who are victims of abuse, for instance: the UN benchmark is one shelter for every 7 500 people.

Or a specialist police officer in every station. Or a dedicated emergency number staffed round the clock. Or simply better data collection: we should not have to rely on researchers extrapolating from mortuary statistics to get a picture of what is happening to women in homes all over our country.