The connection between (social science) knowledge and policy
has been a longstanding debate in public policy (see e.g. Lynn,
1978). Sustainability indicators are at the same time a means to
compile and structure knowledge and to express societal and
political norms and priorities. While this dual role has been
recognized by many scholars writing on the subject, it is less clear
in how far this dual role has been recognized in the design and
implementation of sustainability indicator development processes.
The cases studied suggest that the socio-political dimension
needs as much recognition and consideration in the design
and implementation of a sustainability indicator development
process as the knowledge or substance dimension. This seems not
to be fully recognized yet. Secondly, science-led initiatives do also
‘‘vote’’ on the inclusion and exclusion of indictors or related subclasses;
this voting procedure is not scientific in a positivist sense,
but can be characterised as ‘political’. Thirdly, while in political
processes most actors are representatives of democratically
elected politicians, participation in science-led process is mostly
restricted to ‘‘invited’’ experts (in their fields) and some policymakers
that are recognized experts in relevant issue areas and may
thus not be representative.
The analysis of the cases shows that in practice all allowed a
number of actors from both domains (science and policy) to
participate in indicator development. However, science and policyled
processes had a different bias: in science-led processes there
was, by design, a bias towards the knowledge production
dimension, and less explicit recognition and inclusion of the
normative political dimension in the indicator development
process, particularly also with regard to participation, and thus
representation issues. This is partly due to the research project
character of these processes, which limits possibilities for
Table 8
Cases: selecting and adjusting indicators to changing social and political norms.
SI set (6) Selecting and adjusting sustainability indicators to
social and political norms
EFORWOOD Initial selection based on political selection process of EU SDI
and EU Sustainability Impact Assessment Guidelines,
no adjustment within the lifetime of the 4-year project
SENSOR Initial selection based on EU Sustainability Impact Assessment
Guidelines, but a few indicators representing new issues,
deemed relevant for multifunctional land use,
were added and developed
SEAMLESS The developed tool is generic to allow for adjustment,
but no adjustment was planned within the lifetime
of the project
MCPFE C&I Yes, in the first revision a few indicators were added
and others strengthened to better reflect emerging
political norms
EUROSTAT SDI Initial selection based on political (EU) strategy (mainly
the EU Sustainable Development Strategy). The SDI
set was adjusted to better reflect the revised EU SDS
68 E. Rametsteiner et al. / Ecological Indicators 11 (2011) 61–70
allocating large budgets and time for such purposes. Anecdotal
evidence also suggests that science-led processes (supposedly
focusing more on knowledge creation than social or political norm
building) find it more difficult to involve policy-makers in the
development process than vice versa, and there are questions of
how to get such indicator sets fully accepted and acknowledged as
legitimate by policy-makers. Certain aspects such as e.g. the spatial
scale (e.g. choice of NUTS3 region) may not be really chosen by
scientists, but may be a predetermined and thus influences the
development of the sustainability indicator set in a very specific
way. This raises the question: what is better, a slightly more
accurate but politically less relevant set, or a slightly less accurate
but politically more relevant set? And, from that follows: what is
the appropriate role of scientists in designing indicator sets:
moderators, knowledge brokers, or leaders of development
processes?
With regard to merging different knowledge areas, all processes
have tried to follow the idea of balancing economic, ecological and
social dimensions, but all processes were relatively weak regarding
the participation of experts representing the social dimension.
There is little indication that proper measures were undertaken to
make sure that an envisaged balance is already reflected in the
balanced involvement of respective expertise. None of the scienceled
processes seems to have devised strict science-based selection
processes or rigorous formal rules to base decisions on exclusion or
inclusion of indicators. Most processes have applied more or less
long lists of general principles or selection criteria as ‘‘rules of
thumb’’ to guide decisions, of which many are provided by the
literature. Decisions on the inclusion or exclusion of indicators
were usually consensus-based. Thus, in science-led processes,
participants may decide on indicators outside their specific
domain as informed citizens, i.e. in a political role, expressing
societal norms, rather than in their scientific capacity. This
indicates the central importance of representation and participation
in the indicator development process, which needs to be
properly designed at the outset.
Regarding the merging of different societal and political norms
through the participation of representatives of these different
domains, the main thematic focus of the indicator set determined
participation, and thereby the resulting topical focus of the
indicator sets. Topical focus and participation in turn is likely to
influence the practical recognition and legitimacy of the sets. In
other words, due to the normative nature of sustainability
indicator sets, such sets are more likely recognized as legitimate
by those actors who assume co-ownership, because they have been
involved. The limits of participation of different policy domains
intended or practically achieved in the development process put
likely limits to recognition and use of the resulting set.
All cases studied implicitly or explicitly recognize that the
indicator sets developed are interim results of a continuous
improvement process, particularly with regard to integrating
emerging knowledge. There are also many indications of a broad
recognition of a learning process that took place during development.
Science-led processes, contrary to government-led processes,
have not explicitly provided for possible adaptation to
changing social and political norms and priorities. This is partly
due to the limited time span available in the context of the projects
within which they were developed.
Are science-led sustainability development processes more
accurate and comprehensive with regard to the knowledge they
are able to embody in their indicator sets? Not necessarily, as
government agency experts are also often technical experts with
‘‘scientist’’ standards, and not elected policy-makers. Neither do
government-led sustainability development processes automatically
achieve more policy relevance. It seems that this is only the
case if such processes directly link indicators to concrete policy
strategies and government commitments. One could assert that
the higher the political ‘‘weight’’ of such commitments, the higher
the political relevance (and use) of related indicator sets. This is at
least what a recent comparison of EU SDI and EU structural
indicators across EU Member States would imply (Pu¨lzl et al.,
2007). It found that, while objectives and structural indicators used
in the EU Lisbon Process across Europe are closely aligned, EU and
national SD strategies and related SDI, and – it seems – their actual
use, are much more diverse.
As sustainability indicator development processes are per se
normative, the normative value judgements have to be more
explicitly acknowledged and given more weight by both scientists
and policy-makers; new kinds of methodological choices regarding
participation and representation need to guide development
processes in order to achieve credibility and legitimacy within
society. This may thus not be possible without a wider form of
participation involving citizens or their representatives.