A steady rise has been occurring in the past two centuries in the proportion of civilians being killed in armed conflict, either through violence directly or from conflict-related hunger and disease.
The international community has responded to the calls to protect innocent victims by developing two parallel principles, the ''protection of civilians'' and the ''responsibility to protect''.
Responsibility to protect (R2P) is an internationally agreed doctrine to protect populations from atrocities, and was set down at the 2005 World Summit and reaffirmed in later years in several United Nations Security Council resolutions.
Protection of civilians is based on international humanitarian law, stemming primarily from the fourth Geneva Convention of 1949. Over the past few decades, this norm has developed robustly through the established practices, decisions and procedures of the UN Secretariat, the Security Council, peacekeepers and humanitarian agencies.
By contrast - as last year's Security Council-approved intervention in Libya showed - R2P remains a topic of debate and controversy.
Because protection of civilians is markedly less contentious, its advocates and practitioners fear contagion from the more politicised R2P. So there is a need for a detailed and nuanced explanation of the two principles' commonalities and differences.
On some points, the explanations are straightforward. While protection of civilians, for example, applies to discrete acts of violence against individuals, R2P has a much narrower scope, applying to mass atrocity crimes only.
While peacekeeping operations have an explicit focus on protecting civilians, they can also be important in R2P action. Atrocity crimes are often performed by rebel groups or state-sponsored militias, and peacekeepers can respond to both these factions.
Contrary to common perception, protection of civilians is not restricted to armed conflict, as defined by international humanitarian law. As we classify it, ''narrow protection of civilians'' refers to the obligations of parties to an armed conflict to distinguish between combatants and civilians and to avoid attacking the latter or harming them beyond what is necessary to achieve military objectives. It has precise legal requirements, such as open fighting between two armed forces.
Broad protection of civilians, on the other hand, is a policy framework used by UN and other peacekeepers, the Security Council, the UN Secretariat and humanitarian agencies. These protection actors aim to contribute positively to the protection of civilians in situations of widespread, grave and lawless violence that have not reached the threshold of armed conflict. Syria in late 2011 was such a case, although sometime last year it tipped into full-fledged civil war.
Another common mistake is claiming that protection of civilians is a legal concept, while R2P is political. In making this claim in his 2012 report on protection of civilians, UN Secretary-General Ban Ki-moon contradicted his previous reports on protection of civilians and R2P.
In his later R2P report, Ban returned to the traditional position that, like protection of civilians, R2P is also based on fundamental principles of international, international humanitarian, international human rights and refugee laws.
We agree. R2P and broad protection of civilians are rooted in, but extend beyond, legal principles. The four R2P atrocity crimes - war crimes, crimes against humanity, genocide and ethnic cleansing - have legal definitions in the 1998 Rome Statute (which governs the International Criminal Court) and the 1948 Genocide Convention.
Some R2P duties - for example those that prohibit complicity in genocide by such means as inciting communal hatred - are found in international law. By recourse to the Security Council, even military interventions are made consistent with international law, as distinct from Kosovo-style unilateralism, where NATO used military force against Serbia without Security Council authorisation.
Like R2P, broad protection of civilians draws on the law but extends beyond its strict requirements. The positive duties of peacekeepers to protect civilians are not dictated by international law. So, too, the Security Council has great discretion over the coercive measures it can take to protect civilians and the situations in which it may deploy troops. In 2011, Libya and the Ivory Coast, where the French military and UN peacekeepers ousted Laurent Gbagbo from the presidency, were cases in point.
Moreover, it is mistakenly believed that protection of civilians, unlike R2P, is always impartial, neutral and apolitical. R2P is comparatively more sensitive, as the presence of atrocities implies a perpetrator that may need to be identified and confronted. Even so, the different categories of broad protection of civilian players have different abilities and face different constraints.
The broad protection of civilians used by humanitarian agencies is fully impartial and neutral. But peacekeeping protection of civilians requires the impartial pursuit of the mandate of the peacekeeping force. This may require rejecting neutrality in order to take decisive action against perpetrators, which occurred in the Ivory Coast.
Equally, while respect for sovereignty is a vital element of international peace, in extreme situations Security Council protection of civilians can require (non-neutral) coercive steps to protect civilians - as it did with Resolution 1973 authorising a flight exclusion zone and NATO air assaults in Libya.
The recent UN report criticising its own lack of action in Sri Lanka highlights the need for different protection players to be aware of one another's limits. Humanitarian protection workers can adopt highly apolitical stances to ensure access to those in need. But as the report said: ''The UN's reference to what was 'political' seemed to encompass everything related to the root causes of the crisis and aspects of the conduct of the war.''
In this situation, it was vital to introduce protection actors with the authority and ability to challenge unacceptable behaviour by a UN member country, Sri Lanka. Humanitarian workers cannot do so. Only the Security Council can create and deploy strong protection forces.
Finally, many hold that peacekeepers, humanitarian and human-rights workers may perform specific atrocity-prevention work, but it is better not to refer to these as R2P activities. We agree that in some situations, needless controversy may arise by referring to atrocity-prevention as R2P. The systematic avoidance, however, of R2P language by those engaged in protecting civilians would result in R2P being spoken about only as military intervention.
This stance would neglect efforts to rebuild a country's institutions and provide international help to prevent atrocities. It would also produce a self-fulfilling collapse of R2P into coercive military intervention, ignoring its many contributions to building a nation's capacity to exercise its sovereignty with responsibility.
This is drawn from Enhancing Protection Capacity: Policy Guide to the Responsibility to Protect and the Protection of Civilians, launched in New York on November 19 last year and downloadable free from: http://www.griffith.edu.au/criminology-law/institute-ethics-governance-law/research/responsibility-to-protect-protection-of-civilians-policy-guide.
Dr Hugh Breakey is a postdoctoral fellow at IEGL; Professor Charles Sampford is the director of IEGL; Professor Ramesh Thakur is the director of the Centre for Nuclear Non-proliferation and Disarmament, Australian National University.