Geoscience Reference
In-Depth Information
ment, or more modestly, to initiate the trajectories which would lead to the desired
future (step 4). Interventions are often discussed as part of the trajectory backcasting
step, for example, by identifying targets or obstacles to be overcome (Quist,
2003).
The Quest integrated assessment tool, developed to support the backcasting
process, is composed of a series of linked sectoral submodels. While the submodels
are generally relatively simple, the tool as a whole is much more sophisticated as it
is horizontally integrated across submodels and is driven by a user-friendly interface.
It is described in detail in two recent papers (Tansey et al., 2002, Carmichael et al.,
2005).
This approach is described in detail because it is one of the few attempts to fully
operationalise a participatory integrated assessment model combined with a robust
participatory process. The geographic focus is on a region facing signifi cant popula-
tion growth and development. Forcing these two domains into close proximity
revealed a number of lessons that speak to the wider challenges of making integrated
assessment policy relevant.
First, participatory processes seek to be open with regards to the shape and form
that the future may take. At the extreme, visioning exercises and even backcasting
approaches typically take public values and opinions as sovereign. In a pure partici-
patory application, any future, considered desirable, should be accepted on face
value. Once one introduces a model into the process, the dynamic changes since,
by defi nition, the model identifi es critical thresholds and boundary conditions that
introduce a measure of constraint. It is not technically feasible to create a model
where every combination of every variable is possible. Even if the model is presented
as a heuristic device that generates highly contingent outputs, it can change the
character of the deliberations among participants. An integrated model is a form of
disembodied expertise and it is not possible to negotiate with the model, to check
its assumptions or to learn more about the dynamics it represents. Moreover, even
if the assumptions are correct, the participatory process may generate a version of
the future that is inconsistent with what are understood to be real biogeophysical
and economic constraints. For instance, the group may decide that over two years,
all houses should be converted to high-density, climate-neutral units. While techni-
cally feasible at the outer limits of possibility, this decision would have vast implica-
tions for the rest of the economy. The problem is that group dynamics may create
a very strong commitment to a set of objectives that a model suggests has wider
and negative implications.
Second, with respect to the political process, it is widely understood that deci-
sion-making is not driven by a comprehensive rational analysis. To think of IAMs
simply as more refi ned tools for creating input for reasoned political choices is naïve.
Decisions are made through the alignment of complex networks of political actors.
Scientifi c research is often carefully framed using terms of reference and selective
funding mechanisms. Even the IPCC is subject to close scrutiny, as research on the
political process involved in the preparation of the summary for policymakers (Shaw
and Robinson, 2006) shows. Scientifi c research and models in particular may be
recruited or used for rhetorical purposes. In the translation of knowledge from the
realm of the laboratory into the realm of the political system, contingency and
uncertainty are squeezed out. While more structured processes such as citizen's
juries hold the promise of subjecting expertise to close scrutiny such processes are
still typically treated as inputs to the institutionalised decision-making mechanisms
