|
Some leaps of faith are too great to be plausible or relevant. Some consequences are too remote from their purported causes to be considered important. But how remote is too remote?
The Government of Iraq claimed in the early 1990s that sanctions left it with no money to purchase essential supplies, including water pumps. The UN Security Council was later harshly criticized for the delays and limitations that had been imposed on access to humanitarian goods in Iraq via the “Oil-for-Food” Programme. Critics argued, however, that the Government of Iraq did find funds to build palaces, and that pumps were employed to drain the southern marshes where opponents to the regime lived. Funds and other resources were fungible under control of Iraqis. Surely both of these factors were important, with the Government of Iraq having control over proximal use of available resources, and restrictions in the level of total resources imposed by sanctions as a contributing but more distal cause.
By identifying proximal and distal causes in a chain of events, it may be possible to define an adequate cause—a condition that, if changed, would fundamentally change the outcome. The most proximal adequate cause is that which is sufficient to cause, or prevent from causing, the outcome.
In other words, to define an adequate cause one must examine if the harm would have occurred in the absence of the factor in question. If the absence of sanctions alone would have prevented a rise in child deaths in Iraq, one can argue that sanctions are an adequate cause of the harm. Supporters of this position point to the falling mortality rates in Iraq with the same government in power in the years prior to sanctions. Critics can argue, however, that other countries such as Yugoslavia and Cuba had a decline in income but did not experience a rise in mortality like Iraq; they could argue that a lack of political will or crisis management ability is an adequate cause.
It is seldom possible to identify all possible links in a chain of causation. Especially in social phenomena, the number of possible factors may be too numerous to detail. It is important mainly to identify a few major potential influences only, since the effect of minor influences is likely to be too small to measure. A causal outcome can be defined without understanding all the causal mechanisms involved. It is, however, not possible to identify the causal mechanisms without first identifying what is believed to be a relevant outcome.
A stepbystep approach to developing a causal model, and the associated actions required, is presented in box 2.19 In addition, a simple example of a causal model to identify some of the proximal causes of child malnutrition is presented in figure 4.
The process of constructing the model in figure 4 begins with the question: “What causes child malnutrition?”. Here one is seeking to identify the causes of a single effect of interest (child malnutrition), an effect that can be quantified using an appropriate indicator such as malnutrition rates (percentage of child population).
Several variables can be associated with one another for a causal chain—for example, “poor access to safe water”, and “incidence of preventable disease”. In this way potential, and then likely, causes of child malnutrition are identified. As the causal model is constructed, one can measure the value of each of the causal factors. In this case, there is also a relationship between the two proximal causes: inadequate dietary intake can increase the susceptibility of children to disease, while many normally preventable diseases can in turn result in inadequate dietary intake. For each one of these proximal causes, its causes, in turn, are also identified, tracing back to distal causes. The example shown in figure 4 identifies two levels of causes.
This multilevel approach to causal analysis has also been used by UNICEF. Using a framework of basic, underlying and immediate causes (shown previously in figure 3), UNICEF constructed causal models to identify factors influencing/constraining children’s right to life and education as part of the “situation analysis” of children in Iraq during 20002001. UNICEF defines the three levels of causes as follows:
Immediate causes: such as disease and inadequate nutrition, which directly relate to life, survival and development rights;
Underlying causes: such as the status of household food and nutrition, as well as social services like water and sanitation, health and education, which promote or prevent wellbeing and development;
Basic causes: which relate to issues such as control and distribution of national resources, institutional arrangements and social organization (including the status of women).
UNICEF’s application of the causal modelling approach to the case of Iraq noted:
"The causal analysis approach is particularly helpful in the case of Iraq, where it is important to be able to distinguish those basic causes attributable to the sanctions regime from other basic causes, as well as from underlying and immediate causes. Sanctionsrelated basic causes can only be addressed in the context of an international political resolution of the present situation, and, as such, are not under the control of national authorities responsible, for example, for social services. However, other basic cases related, for instance, to institutional arrangements can be addressed by national authorities if a convincing case is made that these are relevant to children’s survival and development.”20
UNICEF’s causal model using the multilevel causal analysis to assess factors influencing children’s right to survival in Iraq is shown in figure 5.
| 
