While welfare, such as free humanitarian aid, is arguably the sign of a civilised society, it is sometimes accused of creating dependency, undermining sustainable selfsufficiency and demeaning its recipients. The idea that dependency is a bad thing and that free assistance de facto creates dependency not only has long roots in the history of humanitarianism, but also is nourished by the strongly held feelings of those who believe that relief too should be in some way sustainable, linked maybe with a desire to move towards more developmental approaches.
Humanitarian agencies tend to look at the situations of people affected by disasters in terms of their needs and rights. Assessments, for example, use these categories to describe their situation and to formulate responses. Need too is a disputed idea: anyone who has done a needs assessment will know that need is hard to use in a practical way. In itself, need or dependency are not on/off or single-level, but vary in both intensity and over time.
This article suggests a new terminology and a new framework. It seeks to add to the discussion on dependency by borrowing a concept from a field unrelated to humanitarianism, namely dispersed dependencies, an idea formulated by the psychologist George Kelly. In 2005, Paul Harvey and Jeremy Lind put the case that the focus should be, not how to avoid dependency, but how to provide assistance so that those who most need it understand what they are entitled to, and can rely on it as part of their own efforts to survive and recover from crisis. Kelly if Ive understood him right, although he did not address what we call humanitarian concerns himself starts from the position that all humans have their various needs met through a set of dispersed dependencies. Self-sufficiency, or independence, thus means having a full set of dependencies that can be reliably counted upon.
Applied in the humanitarian sphere this is not a revolutionary idea. Harvey and Lind use the concept of interdependency in support of a similar argument to understand the role that aid plays within the multiple forms of interdependency that make up peoples livelihoods, and how these change during crises. A similar idea is that of Vulnerability and Capacity Assessment (VCA) espoused by the IFRC, which is designed to assess the risks that people face in their locality, their different levels of vulnerability to those risks, and the capacities they possess to cope with a hazard and recover from it when it strikes.
The additional point being made here is that there is no such thing as independence in terms of survival, and that therefore dependency should be seen not only as inevitable but as a good thing. Taking this view will enable agencies to approach issues and activities that are most troubling or contentious in a different way, apply a different perspective to them, alter the mind-set in such a way as to offer the possibility of making them less troubling. The value in this idea is that it enables those affected by disasters to be seen in a way that reflects their complicated reality, and what this implies in practice in responding to the crises and disasters people face.
Dependency and coping
Peoples ability to act is a function of the social and economic resources at their command. The extent to which they can purchase or leverage care for themselves through their set of dependencies both formal and informal is what we often call coping. People affected by disasters depend, variously, on themselves, their friends, family and neighbours, hosts, their government, local organisations, international agencies and more. From them they get, if they are lucky, some or all of the material items, space to live, remittance money, food, shelter and water they need to survive, be healthy, live with dignity and in safety, and continue or rebuild their lives. Some dependencies (for instance dependency on humanitarian aid) are stigmatised; others (on community) are valued or idealised. By itself, humanitarian aid cannot create the impossible, that is independence, but it can act to fill out the set of dependencies that people require, and can then work within a framework that transfers dependencies to where they are sustainable, or support those that are sustainable, thus enhancing capabilities.
Why dispersed dependencies?
There seem to be several potential advantages to looking at crisis situations, doing the assessments and making the programme plans through the perspective of the dispersed dependencies of disaster-affected people. This is not to imply that this suggested terminology and concept will either be a panacea or will suddenly make us do a set of things that have not been done before. But they might provide both a conceptual and a practical framework that will increase the likelihood that some of the more enduring problems of humanitarian action will be solved.
First, it will induce a healthy humility in agencies about their place in the lives of people affected by disasters, whether natural or anthropogenic. Logically, behind the concerns about dependency caused by assistance is the idea that this assistance forms a significant part of the coping strategy of the people being assisted. This may be true, and in some circumstances certainly is; indeed, it is what makes assistance worth giving. There is no sense providing a resource that is either meaningless or inappropriate. But looking at the various dispersed dependencies of disasteraffected people will make this clear, and action can be changed, maintained or stopped accordingly.
In addition, help and assistance have to make sense within the culture and practice in which the intervention is taking place. Here, the role of consultation and local communities deciding their own path to survival is critical. To some extent life becomes focused on physical needs and resources to survive, but for people affected part of rebuilding life after a disaster involves community activity and rituals. If we ask them and ourselves about these matters, although they lie beyond our remit, we can try to ensure that they are taken into account in our interventions. Coping with catastrophic personal change, such as bereavement or serious illness, seems to be influenced by what in essence might have stayed the same the need to anchor yourself in some kind of continuity. In the case of large-scale disaster we humanitarians should design and make our interventions in this same light.
An analysis of the set of affected peoples dependencies even if only the physical ones will enable us to see the contributions that are brought with them, that can be reclaimed, that persist through the disaster, and that are created in response to a crisis or disaster. This will, for example, reinforce awareness of the actual roles of local agencies, whether governmental or non-governmental, something that humanitarian actors often do not do well enough. For disaster-affected people, such agencies are part of the context that is beyond [humanitarian agencies] ability to control or improve, and with which we must do our work. Government and community are simultaneously unavoidable partners and constraints for us. None of the above is new as such, but they are among the things that are often found in reviews and evaluations to be missing or lacking. Anything that adds to the pressure to improve performance in action is to be welcomed.
Likewise, the enduring problem of coordination will not be easily solved, as has been proved by the limited success of any process suggested to date. I see the possibility that, if external agencies take on the idea of dispersed dependencies, it will provide a picture within which agencies can seek roles to provide something that people can depend on. This will encourage coordination among external agencies; the role of each will be naturally set within a commonly perceived set of contributions to be made to the set of dependencies of those being assisted.
One of the issues in shaping an assessment into a response involves matching the scale of the response to the perceived need. This is often felt to be supply- rather than demand-driven, and too rarely is it designed as a contribution to meeting needs. An analysis of the gaps among the totality of required dependencies that can be fully or partially filled by an intervention is able to provide a scale of size of packages against a real background. It allows a nuancing of degrees of dependency. It does not imply assistance leading to dependency, but assistance within a context of dispersed dependencies. It gives a realistic meaning to the idea of self-sufficiency, and allows support for choices in either rebuilding previous dependencies or in looking for viable, possibly new, ones. It will support a livelihoods approach, but should ensure that there is no extreme version of self-sufficiency implicit in programme planning and implementation.
Dispersed dependencies and the transition to recovery
In the life of an emergency response there is a theme that rears its head at two significant moments. At the beginning it appears as a claim that an emergency intervention will undermine both self-sufficiency and ongoing development programmes which are predicated on stimulating or relying on self-sufficiency. In the later phase of transition to recovery it reappears as an issue over timing and strategy for exiting from a programme. An original, and repeated, analysis based on the idea that people naturally have a set of ever-shifting dispersed dependencies can lead naturally to shifting interventions to support transitions to recovery.
The need for repeated assessment, as opposed to a oneoff initial assessment and defined programme, has been brought out among other places in the Tsunami Evaluation Coalition Summary Report and the ALNAP Review of Humanitarian Action. There is more than one problem here. But one problem that can be addressed by the idea of dispersed dependencies is that the nature of what is being looked at and looked for is fundamentally different in the turmoil of the acute phase of an emergency and in a time when recovery or rehabilitation is on the cards. This means that the acute phase assessment cannot act as a baseline for later and ongoing assessments. Again, basing the early assessment on a framework of dispersed dependencies enables a continuity that will overcome at least this aspect of the difficulty in justifying and organising repeated assessment.
The overall effect would be to redefine disaster in terms of its effects in radically and rapidly disrupting the patterns of existing (dispersed) dependencies that support people in normal times. Questions remain, of course, not least the relationship between this idea and rights and the implications for the protection approach. None of them seems to me to be fatal to the idea. This is not at odds with the more conventional definition of a disaster as a state or event beyond the coping capacity of the people involved. In fact, it gives meaning and substance to the idea of coping within that definition.
People affected by disasters are suddenly faced with the shaking up of their habitual (dispersed) dependency system and are forced to make sometimes life-critical decisions about how to deal with and incorporate into their strategies possibly unfamiliar potential support (the inverse of the dependency idea). The role of the humanitarian system should be to support them in doing this and in the difficult choices that they have to make. Dignity, an essential but intangible aspect of humanitarian aid, in part comes through being allowed to manage your dependencies. This can break the link between being in need of aid on the one hand, and powerlessness, incapacity and neediness on the other.
Maurice Herson is Editor of Forced Migration Review (http://www.fmreview.org).
 1 In Kellys field, it refers to the set of individuals relationships, where they have practical, emotional, social, physical, etc. dependencies. Personal independence is about having a viable set of dispersed dependencies dependency is neither a problem nor abnormal. (No man is an island, as the poet said.)
 John Cosgrave and Maurice Herson, Perceptions of Crisis and Response: A Synthesis of Evaluations of the Response to the 2005 Pakistan Earthquake, in ALNAP Review of Humanitarian Action, 2008 (London: ALNAP, 2008), p. 214.