What is a Narrative?

According to T.F.H. Allen a narrative is the result of a series of scaling operations over the perception of a given event used for identifying: (i) relevant agents, (ii) a given scale of analysis, and (iii) a direction of causality, a combination providing an explanation for the event. Narratives are essential for humans because:

1. Narratives provide anticipation, but a tricky anticipation

The representation of relevant events based on the choice of a given narrative makes it possible to describe dynamic relations – i.e. the scaling makes it possible to define a “before” and an “after” and a causal link in the plot of perceived events. This explains why models need (and depend on) a pre-analytical choice of a narrative about the events to be modeled.

2. Narratives make it possible to make sense of our perception of the reality

Narratives represent a heuristic solution adopted by humans to avoid impredicativity. When dealing with complex systems it is unavoidable to find contrasting perceptions in which: A -> B, but also B -> A (i.e. chicken-egg paradox).

Impredicativity generates a bifurcation in the definition of:

  • what should be considered a relevant agent (dependent or independent variable in the model). This choice depends on the point of view used for the perception. Governments rule on citizens over a time duration of a year, whereas citizens rule over the government at election time (over a time duration of 10 years);
  • what should be considered the right scale to be used for analysis (it blurs the definition of “before” and “after”) – the famous chicken-egg dilemma!; and
  • what should be considered the right direction of causality used to provide an explanation over the relations agents/phenomena perceived in the external world – Are farmers needed to reproduce rural communities or rural communities needed to reproduce farmers? The same impredicative relation applies to the relation between urban and rural communities.
3. Narratives cannot handle the trade-off between efficiency and adaptability

The use of narratives does not solve the impasse faced when trying to deal with the well-known and non-quantifiable trade-off between “efficiency” and “adaptability”. When optimizing efficiency (what is more efficient right now, under the ceteris paribus assumption) we systematically reduce the diversity of possible solutions: those described as “not optimal” by the chosen formalization, which is based on a given set of relevant criteria and the existing perception of external boundary conditions. Because of this forced trade-offs, more efficient systems become less capable of adapting (they lose their potentiality of expressing different functions) when the definition of relevant criteria of performance and/or the constraints determined by boundary conditions will change. When addressing the trade-offs between efficiency and adaptability (resilience!!!!) we cannot focus on the detailed indications given by models or data. Looking only at the results of models carries the risk of missing the presence of elephant(s) in the room(s). This is why, the basic goal of MAGIC is to develop an approach that will allow us to check the usefulness of the narratives that are behind the choice of models and data used as “evidence” for policy.

The usefulness of a narrative depends on the purpose and the value judgment of the agent endorsing and using the narrative (see What is a story-teller?).

Allen, T.F.H. and T.W. Hoekstra 1992. Toward a Unified Ecology. New York: Columbia University Press.
Allen, T.F.H., Starr, T., 1982. Hierarchy: Perspectives for Ecological Complexity. University of Chicago Press, Chicago.

For more information: M. Giampietro (2017): Structuring the perception (qualitative) and representation (quantitative) of the nexus with new concepts and narratives. In: Report on Nexus Security using Quantitative Story-Telling. MAGIC (H2020–GA 689669) Project Deliverable 4.1.