Next generation Sustainability Governance Design Assignment

Collaborative governance and adaptive management: Lessons from California’s CALFED Water Program

Giorgos Kallis a,*, Michael Kiparsky b, Richard Norgaard b

a ICTA, Universidad Autonoma de Barcelona, ETSE, QC/3095, 08193 Bellatera, Barcelona, Spain bEnergy and Resources Group, University of California at Berkeley, 310 Barrows Hall, Berkeley, CA, 94720-3050, USA

e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 2 ( 2 0 0 9 ) 6 3 1 – 6 4 3

a r t i c l e i n f o

Keywords:

Collaboration

Adaptive management

Governance

Water policy

California

a b s t r a c t

Both for its technological and institutional innovations and for its history of conflicts,

California’s water system has been one of the most observed in the world. This article and

this Special Issue on the CALFED Bay-Delta Program continue in this tradition. CALFED is

likely the most ambitious experiment in collaborative environmental policy and adaptive

management the world has seen to date. This Issue moves beyond the celebratory tone of

other analyses of collaborative, adaptive management and looks closer into how collabora-

tive networks work to produce innovation, and more importantly to reflect also on their

inherent contradictions, limitations and ‘‘dark sides’’. While collaborative governance

enhances mutual understandings and can be a source of innovation, it appears ill-suited

to resolve alone the distributive dilemmas at the core of many water – and other environ-

mental – conflicts. A lacuna in existing research concerns the institutional design of

effective boundaries and linkages between democratic politics, legitimate authority, and

adaptive governance, i.e. the mix of institutions that can provide sufficient responsibility,

accountability and democratic legitimacy, without choking off the self-organizing interac-

tion, shared learning, and communication that is at the heart of collaboration. A painful

realization in the Delta is that environmental conservation and further growth may be

fundamentally at odds; efficient win–win solutions, institutional or technological, seem

insufficient to satisfy the competing demands posed upon the system. Radical decisions and

changes might be necessary, but they seem unlikely under current institutional arrange-

ments and political conditions.

# 2009 Elsevier Ltd. All rights reserved.

avai lab le at www.sc iencedi rect .com

journal homepage: www.elsevier.com/locate/envsci

1. Introduction

California’s water system has for decades stood as not only

one of the largest and most complex in the world, but also

one of the most innovative. As such, it has also been one of

the most observed water systems. This Special Issue

continues the tradition of studying and learning from

California’s innovations, with a multi-disciplinary collection

of articles and commentaries focused on the CALFED Bay-

Delta Program, perhaps the most ambitious experiment in

* Corresponding author. Tel.: +34 93 581 3749. E-mail address: [email protected] (G. Kallis).

1462-9011/$ – see front matter # 2009 Elsevier Ltd. All rights reserved doi:10.1016/j.envsci.2009.07.002

collaborative policy and adaptive management the world

has seen to date.

Water has long been a critical resource in California, one of

the world’s largest economies. The Bay-Delta (Fig. 1) is an

estuary comprised of San Francisco Bay (hereafter ‘‘the Bay’’)

and Sacramento–San Joaquin Delta (hereafter, ‘‘the Delta’’).

The Delta is both a fragile, valued ecosystem and the water

‘hub’ of California through which 22 million people, two thirds

of the state’s population, receive at least some of their

drinking water. Its management also directly impacts the

.

Fig. 1 – Regions of California, with detail of the Sacramento-San Joaquin Delta. Figure courtesy of California Department of

Water Resources.

e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 2 ( 2 0 0 9 ) 6 3 1 – 6 4 3632

condition of habitat in the Bay. The Delta has been the central

battlefield in California’s ‘‘water wars’’. Interventions to

transform it have pitted north against south, locals and

environmentalists against cities and farmers (Hundley, 2001).

In 1994 over 25 federal and state agencies and representa-

tives of more than 30 major stakeholder groups and local

agencies agreed to collaborate in an integrated program of

restoration and management of the Bay-Delta. The CALFED

program, as it came to be known, was celebrated as the end of

California’s water wars (Rieke, 1996). $3 billion were spent on

restoration, research, and water banking between 1994 and

2006. CALFED is probably ‘‘the world’s most extensive – and

expensive – water management program’’ (CALFED, 2000, p. 1).

$1 billion alone went to what is ‘‘the largest program of

environmental restoration in American history’’ (Hundley,

2001). CALFED was praised as ‘‘a leading edge experiment’’ in

collaborative planning (Innes et al., 2006, 2007), a new model of

environmental regulation (Freeman and Farber, 2005) and an

exemplar of adaptive management (Hundley, 2001). Yet a

general discontentment with the program’s management,

coupled with its failure to achieve in the short-term its stated

goals, led to its eventual dissolution by 2007 (Fig. 2).

What happened in the Bay-Delta concerns the rest of the

world. Collaborative, adaptive governance based on inter-

agency integration and stakeholder participation is the new

paradigm for managing environmental problems (Lemos and

Agrawal, 2006; Folke et al., 2005). These ideas underpin major

policy initiatives elsewhere in the world, such as the EU Water

Framework Directive (Kallis and Butler, 2001). This Special

Issue aims to shed light on the attributes of effective

environmental governance, but also on its limitations and

contradictions. Any attempt to understand a program of such

immense scale and complexity might be likened to blind men

and women groping at an elephant; a unified, complete picture

Fig. 2 – Some key issues and conflicts in the Delta. Issues in the California Delta are myriad, interconnected, and complex.

This brief description offers some contextual information for readers of this Special Issue but does not do justice to the

complexity of problems and differing viewpoints on each of the above. See Healey et al. (2008) for a recent assessment of

the scientific issues in the system, and other articles in this issue for entry points to ongoing Delta debates.

e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 2 ( 2 0 0 9 ) 6 3 1 – 6 4 3 633

is almost impossible to construct. This Issue brings together

the scholarship of a multi-disciplinary, multi-perspective

group of researchers and practitioners in the hopes that the

whole of our individual observations will amount to more than

the sum of its component parts. Our findings speak both to the

international academic community concerned with environ-

mental governance and to the California water policy

community. We hope that our collective insights add some-

thing to the search for viable solutions to the water conflicts

centered around the Bay-Delta, and can inform efforts to

tackle other complex environmental management proble-

ms.This article opens the Special Issue with a literature review

that synthesizes existing studies on CALFED with the invited

contributions in this issue. Section 2 gives a brief history of the

evolution of institutions in the Bay-Delta. Section 3 revisits the

literature on collaboration and adaptive governance, much of

it informed by past studies of CALFED. Section 4 recaps the key

insights of the contributions in this issue concerning the

procedural attributes of successful collaboration. Section 5 in

turn identifies potential limitations, particularly in relation to

questions of political economy, justice and the still prevalent

notion that humans can ultimately control complex socio-

ecosystems. We conclude in Section 6 reflecting on the

aftermath of CALFED and its lessons for similar initiatives

in different scales and locations.

2. The California Bay-Delta and CALFED

The Delta drains 60% of California’s runoff, all of the water

flowing into the Central Valley with most of that from Sierra

Nevada mountain range. It is the hub of California’s complex

water distribution system, consisting of the Federal Central

Valley Project (CVP) and California’s State Water Project (SWP).

The projects consist of major reservoirs that store water that

falls as precipitation mainly as snow in winter. This water is

conveyed through a network of rivers and built infrastructure

from the water-rich northern and eastern areas of the state to

the population and agricultural demand centers in the south.

Much of this water is pumped through the Delta.

The lineage of Delta conflicts can be traced back to the 19th

century when levees were built on the peat soils to reclaim

land for farming (Mount and Twiss, 2005) and gold miners and

farmers fought over the sediment flowing through mining

operations in the Sierra foothills. With the construction and

expansion of CVP and SWP in 20th century, conflicts centered

around water allocation, environmental protection and water

quality leading to the founding of the State Water Resources

Control Board (SWRCB) in 1967 (Norgaard et al., 2009;

Hanemann and Dyckman, 2009; Hundley, 2001). A major

political milestone was voters’ approval of a statewide ballot

in 1982 against a Peripheral Canal, that would by-pass and

e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 2 ( 2 0 0 9 ) 6 3 1 – 6 4 3634

convey water around the Delta to the south (Hanemann and

Dyckman, 2009; Gottlieb, 1988). The argument of the environ-

mental groups and Delta communities and farmers that

initiated the referendum was that the Canal would facilitate

higher exports from north to south, impacting fisheries and

ecosystems in the Bay-Delta, leaving less water for local uses

and increasing saline intrusion from the Bay, arguments that

resonated particularly with locals and Northern Californians

(Hanemann and Dyckman, 2009 for a more nuanced analysis).

In the years after the ballot, attention shifted from increasing

water supply to improving water quality (Gottlieb, 1988).

Diffuse sources of pollution from agriculture, automobiles and

abandoned gold mines contaminated the Delta and Bay. In dry

seasons, seawater intruded the Delta as pumps exported more

freshwater. In 1986 the California Court of Appeals made an

unprecedented ruling, the ‘Racanelli decision’, which in effect

ordered the SWRCB to broaden its mandate to protect all

beneficial uses of water, not restricted as before to water

rights. This in effect shifted some of the onus for water quality

and environmental protection to senior water users and

strengthened environmental protections (see Hanemann and

Dyckman, 2009). As state authorities and the SWRCB were

reluctant to impose such limits on agricultural and urban

water users, Federal authority moved to fill the void. The EPA

under the Clinton Administration ordered the State to reduce

water exports from the Delta and bring salinity levels within

the standards of the Clean Water Act. Delta smelt and winter

run Chinook salmon were listed as endangered under the

Endangered Species Act (ESA), perhaps the most powerful

environmental law in the U.S. Then during the 1992 drought

that threatened agricultural and urban supplies, Federal fish

agencies ordered the projects’ pumps halted to reduce the

uptake of endangered fish (Connick and Innes, 2003). In the

same year the Congress passed the Central Valley Project

Improvement Act, dedicating substantial environmental flows

and funds for restoration in watersheds above the Delta.

Around this time several alternative processes arose in

attempts to head off the growing crisis. First, the San Francisco

Estuary Project (SFEP) brought agricultural, urban, business,

and development interests together to develop a plan for

restoration and management of the Bay (Connick, 2003).

Second, the SFEP spawned a ‘‘Three-way process’’ in which

agricultural, urban, and environmental groups met informally

in attempts to reach agreements on flows themselves

(Connick, 2003; Innes et al., 2007). Third, four Federal agencies

(Environmental Protection Agency, Fish and Wildlife Service,

National Marine Fisheries Service, and US Bureau of Reclama-

tion) pledged to cooperate on Bay-Delta actions. While none of

these processes alone resulted in an overarching solution to

the problems in the Bay-Delta, they did generate capacity

building and institutional learning.

As environmental stresses in the Delta increased, attempts

to forge solutions built towards a key turning point: in 1994,

the Bay-Delta Accord established CALFED’s collaborative

process by bringing stakeholders together in a new forum.

However, it did not describe how its goals would be

accomplished, assign institutional responsibilities, or seek

its own implementation and enforcement authority. State and

federal agencies and selected stakeholders established a set of

interim measures covering export limits, operational flex-

ibility to comply with the ESA, and measures to improve

environmental conditions in the Bay-Delta. The mandate was

to come up with an integrated program for the Bay-Delta. This

eventually took the form of the Record of Decision (ROD), a

requirement for the Environmental Impact Statement process

of the program. CALFED got financial support from state and

federal governments, in-kind support from agency staff

dedicated to the effort, as well as three bonds passed under

the initiatives of stakeholders, with a total of $3 billion

(Hundley, 2001).

In this first phase of CALFED between the Accord and ROD

(Fig. 3), the process was led by a high-level Policy Group with

heads of state agencies and high-level officials from federal

agencies, supported by a Management Team of deputy

directors charged to implement their decisions. Formal public

input came from a Bay-Delta Advisory Committee (BDAC)

made up from stakeholders from different interest groups

(Innes et al., 2006). However, most of the work and initiative

was taken up by ad hoc interagency, stakeholder and mixed

groups under the umbrella of the Policy Group or BDAC that

developed agreements and action plans for particular issues.

The Record of Decision was signed in 2000 by 24 State and

Federal agencies and included – among eleven programmatic

areas – action plans for levee system integrity, ecosystem

restoration, drinking water quality, and water supply relia-

bility, and an innovative Science Program (ROD, CALFED Bay-

Delta Program, 2000). In effect, the ROD formalized an

agreement by which environmental restoration and limited

releases of environmental water would be exchanged for

continued exports. The effects of this ‘‘soft’’ option were to be

evaluated after seven years, while the possibility of a (smaller)

Peripheral Canal would be studied (Hundley, 2001). The ROD

was notable for its scope and ambition, and for its lack of

specificity; it was more an ‘‘agreement about heuristics for

continuing to work together’’ (Innes et al., 2007, p. 204). The

Legislature decided to set up an independent California Bay

Delta Authority (CBDA) to oversee the program, composed of

Governor appointed regional representatives and selected

agencies heads (LHC, 2005). Much of the rest of the CALFED

program remained structured around working sub-commit-

tees, such as the Environmental Justice committee (Shilling

et al., 2009) and programs and work-groups such as the

Science Program and the Ecosystem Restoration program

(Norgaard et al., 2009; Taylor and Short, 2009) or the EWA

(Lejano and Ingram, 2009), that are discussed in this Special

Issue.

Whether CALFED failed or succeeded in its mission is

subject to debate, as we discuss below. An evaluation ordered

by the Governor towards the end of the seven-year period

concluded that the program had largely failed to achieve its

goals, particularly those of reversing declining species

populations and improving levee stability. The CBDA was

disbanded, downsized, and passed to the Resources Agency,

with only the Science Program surviving intact.

As this Issue goes to press, the mantle of comprehensive

Bay-Delta institutional arrangement remains unclaimed. The

Governor ordered a Delta Vision Blue Ribbon Task Force to

produce a new vision for the Delta and a strategic plan, and

more emphasis has been given to an existing agency, the Bay

Conservation and Development Commission. Independent of

Fig. 3 – CALFED Timeline.

e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 2 ( 2 0 0 9 ) 6 3 1 – 6 4 3 635

these processes, however, the Governor has also ordered

studies to begin on the feasibility of a new Peripheral Canal,

while avoiding the critical issue of the institutional coordina-

tion of the 200 federal and state agencies that share authority

in the Delta (Reilly, 2009). It remains to be seen whether these

or other efforts will emerge as successors to CALFED’s

attempts at integrative management.

3. Collaboration and adaptive governance: key concepts

CALFED can be characterized as a collaborative, adaptive,

governance process. In this section, we briefly introduce the

literature on each of these concepts to ground the interested

reader in some of the relevant academic theory.

3.1. Environmental governance

Environmental governance refers to processes of negotiation,

coordination and collaboration between state agencies,

private actors and non-governmental organizations directed

to the joint realization and implementation of a plan

addressing an environmental problem (Jessop, 1998).1 Govern-

1 The term governance is elsewhere used more generally to denote any governing institutional arrangement. For example other articles in this issue refer to ‘‘Delta Governance’’, meaning the institutional organization that will govern the Bay-Delta. In this article we use a scholarly definition of the term referring to heterarchic, polyarchic self-organizing forms of government.

ance processes are polycentric, heterarchic and self-organiz-

ing. The definition and the response to a problem are under

negotiation, provisional actions emerging through the infor-

mal interaction between participants, continuously updated

and modified as new information and new interests are

brought to the negotiating table (Jessop, 1998). CALFED,

especially in Phase 1 (Fig. 3) has been cited as an exemplar

of good governance: policymakers, stakeholders and scientists

collaborated in ad hoc, self-organizing work-groups under a

fluid institutional structure and produced innovative agree-

ments that surpassed long-standing stalemates (Innes et al.,

2007).

CALFED is a particular type of environmental governance.

First, CALFED is a large-scale process (Heikkila and Gerlak,

2005), as compared to smaller-scale, decentralized governance

of common-pool resources, where collaboration may be

somewhat easier (e.g. Ostrom, 1990).

Second, CALFED is a multi-scale institution with nested,

smaller-scale governance and negotiation processes, such as

those of working groups and programs described in some of

the articles of this Issue. The contributions to this issue fall

out roughly into two broad scale-based categories. Many of

the contributions focus on the smaller scale, finer grained

analysis of a discrete segment or sub-process, such as the

WUE or EWA (Taylor and Short, 2009; Lejano and Ingram,

2009; Fuller, 2009). This level of focus enables insights into

the social, interpersonal, and management-level aspects

that enable innovation and generation of novel solutions.

Other contributions step back and look at a larger spatial/

temporal/institutional scale, coarser grained analysis of the

‘‘entire’’ CALFED process, enabling conclusions about the

e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 2 ( 2 0 0 9 ) 6 3 1 – 6 4 3636

importance of institutional design, boundaries, history, and

legal aspects in the evolution of such efforts (Hanemann and

Dyckman, 2009; Norgaard et al., 2009; Owen, 2009; Shilling

et al., 2009). Both approaches are complementary: although

they may not always be easy to reconcile, we embrace

the tensions resulting from this juxtaposition of scales of

analysis.

Third, CALFED is a cross-scale governance process, i.e. it

extends across different levels of social and institutional

aggregation (Lemos and Agrawal, 2006; Dengler, 2007).

CALFED brought together agencies and actors from the

federal, state and local levels, with very different spatial

and functional boundaries. It is precisely contemporary large-

scale, ‘‘wicked problems’’, such as the management of land,

water and species in the Bay-Delta, whose boundaries are

mismatched with the boundaries and assets of existing

administrative jurisdictions, that require flexible, integrative,

cross-scale institutional arrangements such as CALFED

(Kettl, 2006).

3.2. Adaptive management

Adaptive management (AM) is a response to the realization

that because managed ecosystems are dynamic and unpre-

dictable it is difficult to predict and control their behavior

(Gunderson and Light, 2006; Holling, 1978). In AM policies

become hypotheses and management actions experiments to

test these hypotheses (Folke et al., 2005). AM requires

continuous monitoring, evaluation and adjustment of poli-

cies. Networks, polycentric governance and collaboration are

pre-requisites for good adaptive management (Folke et al.,

2005). Scientists assume a ‘‘new role’’ in an AM context,

shifting from experts to ‘‘one of several actors in the learning

and knowledge generation process’’ (Folke et al., 2005, p. 445).

‘‘Adaptive co-management’’ refers to collaborative programs

of adaptation, whereby policy design and evaluation is a joint

process. ‘‘Adaptive governance’’ refers to the social and

institutional arrangements that provide an organizing frame-

work for adaptive management (Dietz et al., 2003).

CALFED was implicitly, if not explicitly, a grand AM

experiment. The Accord (and later the ROD) struck a

compromise following the Peripheral Canal and Salinity

controversies which was to ‘‘wait and see. . .act and learn’’,

investing meanwhile in a mix of non-conflictive options from

competing proposals, studying the system more and creating

an informed agreement later (Hundley, 2001). Specific sub-

programs of CALFED, such as the Ecosystem Restoration

Program and the Science Program were explicitly built on AM

principles (Taylor and Short, 2009).

Gunderson and Light (2006) argue that AM is not just trial

and error management or management by objective with

evaluation and updating. AM requires large-scale experi-

ments and an acceptance that failures occur but they offer

learning opportunities. As in the Comprehensive Everglades

Restoration Plan, the other large-scale AM restoration

program in the U.S., in CALFED large-scale experimentation,

such as significantly reducing pumping and exports, was

limited. Furthermore, the ‘‘scientific management approach’’

which requires certitude prior to action (Gunderson and

Light, 2006) remained strong throughout; policymakers had

little patience for bad news. While it may be tempting to

conclude that CALFED did not measure up to the AM ideal and

leave it there, in this article we want to ask how and why

AM worked the way it did in the real-world context of the

Bay-Delta and California water policy, and thus to engage

more with the external limitations and the inherent contra-

dictions of AM.

3.3. Collaboration

Collaboration is at the heart of adaptive governance. Colla-

boration means to co-labor, to work together (O’Leary et al.,

2006). It is not merely power-brokerage, i.e. trading among

predefined interests to find an optimal point of agreement

(Fuller, 2009). Engagement and interaction may create new

value and mutual social learning. Collaboration among

partners in CALFED is said to have reframed a struggle over

water users’ entitlements to the collective question of ‘‘what

do we want this watershed to do?’’ (Freeman and Farber, 2005,

p. 3). Such reframing allows new ideas to emerge that were not

part of a polarized solution spectrum. An oft-mentioned

example in this respect is the Environmental Water Account

(EWA) (Innes et al., 2007; Freeman and Farber, 2005; Ingram

and Fraser, 2006; Lejano and Ingram, 2009). In the EWA

environmental and water agencies trade water for fish with

water for drinking and agriculture in real time. Innovative

ideas like the EWA, some scholars have argued, are most

readily conceived through informal interaction between

agencies and stakeholders, such as those in the CALFED

working groups (Innes et al., 2007; Ingram and Fraser, 2006).

Such interaction not only produces innovation but also creates

a ‘‘cascade of changes in attitudes, behaviors and actions’’ and

‘‘social and political capital’’ with long-term positive effects

(Connick and Innes, 2003).

However in this Issue we want to go one step further, not

only in terms of understanding how collaboration works, but

also engaging with its ‘‘dark side’’ and shortcomings (McGuire,

2006). Because collaboration is new, or because it produces

new results, it does not follow that in and of itself it must be

desirable (McGuire, 2006).

Whereas distinctions between collaborative, adaptive

governance and hierarchical state regulation and competi-

tive markets are often emphasized, in the real world these

three forms necessarily coexist and depend on one another

(Jessop, 1998). CALFED for example did not replace but

incorporated conventional regulatory agency programs, in

the process allowing them to develop new connections and

innovations. Governance needs State forms of governing.

Court decisions for example formed the background

entitlements with which CALFED partners sit at the

negotiating table (Freeman and Farber, 2005). State support,

financial and symbolic, was crucial and so were state

assurances that agreements will be implemented. Further-

more, the State offers a governance process the democratic

legitimacy that it otherwise lacks given the ad hoc selection

of participants. The downfall of this is that inversely,

governance suffers from the shortcomings of State admin-

istration and it is vulnerable to external, political changes

(Thompson and Perry, 2006). Many of the shortcomings of

CALFED, for example, have been related to general problems

e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 2 ( 2 0 0 9 ) 6 3 1 – 6 4 3 637

of public administration in the participating agencies and

CBDA, such as understaffing, budget management pro-

cesses, competition between state and federal agencies or

entrenched agency mentalities (Lurie, 2004). Political

changes such as the election of the George W. Bush

administration after the signing of the ROD and California’s

budget crisis shortly thereafter also undermined CALFED

(LHC, 2005). However, such changes cannot be viewed as

unexpected aberrations that derailed an otherwise success-

ful governance program; in the real world, governments

change and crises happen. Nor can government interven-

tions be seen as ‘‘messing’’ with an otherwise innocuous

governance process; governance needs the State. Govern-

ance processes therefore have to be studied within their

real-world institutional context, and in their real, messy,

hybrid form.

4. When and how does collaborative governance and adaptive management work

Networked, collaborative governance arrangements are cru-

cial for a culture and practice of adaptive experimentation

(Folke et al., 2005; Gunderson and Light, 2006). Favorable

conditions for their emergence include: an impasse which

makes warring factions ready to negotiate alternatives (i.e.

‘‘fail their way into collaboration’’, Bryson et al., 2006); a

relative balance of legal, economic, and/or political power

(Duane, 1997); pre-existing social capital and networks;

stakeholders with the resources and expertise necessary to

generate new solutions; political mandate, pressure and

support; and the presence of – or prospect of access to –

external financial resources that would not otherwise be

available to participants (Freeman and Farber, 2005; Bryson

et al., 2006). These conditions were largely met in CALFED

(Innes et al., 2007; Freeman and Farber, 2005). The ballot defeat

of the Peripheral Canal and the series of legal decisions

empowered environmental groups and created legal and

political impasses. Expertise and scientific knowledge were

distributed beyond agencies and Universities. Stakeholders had

already started networking in the San Francisco Estuary Project

and the Three-way process (Connick, 2003). And federal and

state leaders, most notably Interior Secretary Bruce Babbitt of

the Clinton administration, pushedtheprocessand lubricated it

with federal subsidies and state bonds.

If these conditions bring stakeholders to the table for

negotiation, they alone are not sufficient to create successful

collaborations and partnerships (Fuller, 2009). Innes and

Booher (1999) and Bryson et al. (2006) identify several

important procedural attributes for effective collaboration

such as: the presence of shared practical tasks; initial

agreements; a reliance on self-organization rather than an

externally imposed structure; the use of high-quality, agreed-

upon information sources; proceeding with agreements when

there is overwhelming support; external legitimacy of the

process; resources and commitment to equalize power

differences between participants; continuous trust-building

activities, and genuine engagement in productive dialogue.

The contributions in this Issue elaborate further how and

when collaboration works.

4.1. Looking inside collaborative processes

Contributors heed Agranoff’s (2006, p. 56) call to ‘‘go beyond

heralding the importance of collaborations to look inside their

operations.’’ They engage with the question of how and when

innovative agreements result, looking at processes and

working groups within CALFED that produced breakthrough

results, and others that clearly failed (Lejano and Ingram, 2009;

Fuller, 2009). Contributions in this issue delve deeper into the

question of how collaborative processes work, training a

magnifying lens on the mechanics of sub-processes within the

larger CALFED program. Their common starting point is that

‘‘it is the shared learning process that is critical’’ (Norgaard

et al., 2009).

A critical institutional avenue towards encouraging shared

learning is the creation of boundary organizations. These refer

to the institutionalized forums where different knowledge and

stakeholders work together to bridge the gaps between

disparate frames and viewpoints. The Science Program (Taylor

and Short, 2009; Norgaard et al., 2009) or the Environmental

Water Account (Lejano and Ingram, 2009) served as boundary

organizations. They provided opportunities for direct, perso-

nal and sustained engagement of scientists and stakeholders,

facilitating shifts in concepts and the emergence of new

language to talk about problems and solutions (Taylor and

Short, 2009).

Within such boundary organizations, boundary objects

are used to develop a shared language—an ‘‘inter-language’’

in Fuller’s (2009) terms. Boundary objects are ‘‘artifacts that

individuals work with. . .that cross disciplinary or cultural

barriers’’ (Carlisle, 2002:446, as cited in Fuller, 2009), such as

models, maps, reports, spreadsheets or power point pre-

sentations, or even the very conferences and workshops

that create a space for shared interaction. Boundary objects

offer stakeholders a new vocabulary to talk about problems

and a platform for modifying and re-organizing concepts in

a way that is acceptable from all perspectives (Fuller, 2009;

Lejano and Ingram, 2009). For example, in the EWA, games

and modeling simulations of pumping scenarios and their

impacts on fish, allowed stakeholders to get a grasp of what

water trade meant and offered a base for negotiation and

agreement (Hudgik and Arch, 2003; Innes et al., 2007).

Identifying commonalities between CALFED and global

scientific assessments, Norgaard et al. (2009) underscore

how shared language can take the form of a new meta-

model or the complementary use of multiple analytical

models with different scales or functions as a way to allow

participants to communicate across disciplinary perspec-

tives.

However, as Lejano and Ingram (2009) show, the narratives

and perspectives of different stakeholders are not reconciled

and integrated just through the creation and use of a master

frame. It is in the conversation, translation and exchange of

different knowledges, i.e. the dialectic juxtaposition of

concepts, that ‘‘magic occurs’’, not in the mere combination

of the knowledge stored by each camp (Lejano and Ingram,

2009, p. 4).

Yet merely adding a boundary organization or object in

the policy mix does not suffice. Whereas the EWA or the

Agricultural Water Use Efficiency Committee succeeded, the

e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 2 ( 2 0 0 9 ) 6 3 1 – 6 4 3638

Water Use Efficiency (WUE) Program or previous attempts to

agree on agricultural efficiency standards, failed (Fuller,

2009; Lejano and Ingram, 2009). Fuller (2009) explains

differences between success and failure focusing on the

management of the collaborative process. He documents the

crucial role of professional facilitators in synthesizing ideas

and inventing the new words that constitute the group’s

inter-language. Maintaining a relationship between the

negotiators and their constituencies is crucial for grounding

agreements in the political realities of the situation. A

question however, is not only how can constituencies feel

part of agreements, but how they can become part of the

new inter-language developed between those active in

boundary work. Fuller emphasizes the importance of a

‘‘safe space’’ of closed doors, whereby stakeholders can

negotiate freed from the symbolic roles they have to

maintain in public arenas (also Innes et al., 2007). Yet

‘‘closed doors’’ pose a trade-off: they offer a safe environ-

ment for negotiation and agreement, but at the risk of

estranging outsiders. In the case of the Agricultural

committee, this was somewhat managed by the continuous,

almost real-time, communication between those in and

those out of the room (Fuller, 2009). The danger of a growing

disjuncture between those engaged in partnership and those

whose interests are being represented always looms and can

eventually undermine the legitimacy of a partnership

(Jessop, 1998). And whereas some degree of isolation is

necessary in order to stabilize a partnership and facilitate

focused and timely action, it is also likely to act as a barrier

to the recruitment of new partners (Jessop, 1998).

4.2. Institutionalizing collaboration

Lejano and Ingram (2009) locate instead the difference of

successful from unsuccessful processes to institutional

design. The institutional set-up facilitates or hinders

networking and the encounters necessary for the creation

of an inter-language (Lejano and Ingram, 2009). WUE was

meant to be a boundary organization, but failed because its

regulatory design in the form of a centralized, coordinating

body implementing and overseeing a set of predefined ‘‘Best

Management Practices’’ did not provide opportunities for

communication, mediation and translation of information

between stakeholders (Lejano and Ingram, 2009). Perhaps

counter-intuitively, new institutions may have better

results as boundary organizations, as they do not bring

the ‘baggage’ of familiar way of doing things and thus are

more conducive to new ideas (Lejano and Ingram, 2009).

Furthermore, the important feature of design may not be

so much its form, but its practice; EWA ostensibly a

market design, functioned well because it worked in a

non-market way, i.e. through personal interactions and

information not bounded by a pricing signal (Lejano and

Ingram, 2009).

This insight from smaller-scale processes within CALFED

echoes the debate about the institutional form of the

program as a whole. The issue there was whether CALFED

after the ROD agreement should be governed as a program by

a dedicated agency or that instead it should continue to be

more of a voluntary, fluid structure of working group

exchange, coordinated by a high-level Policy Group. The

argument in favor of a centralized, coordinating body was

that it was necessary to offer assurances for the implemen-

tation of the agreements and political responsibility and

accountability for the vast amounts of money managed. The

counterargument was that the institutionalization of

CALFED in a coordinating regulatory agency, would choke

off the innovativeness of ad hoc partnerships, increase

bureaucracy, reproduce the problems of agency-based public

administration and create competition between existing

agencies and the new agency. While some argue that this is

precisely what happened with CBDA (Innes et al., 2007),

others argue that the problem was precisely the opposite, i.e.

that CBDA did not have enough power over the agencies in

terms of managing the funds and assuring implementation

(Bobker, 2009; LHC, 2005).

This tension between governability and accountability on

the one hand and flexibility, creativity and adaptability on

the other, has been noted elsewhere (Jessop, 1998). Internal

accountability within the process, i.e. each partner being

watched by all others (Freeman and Farber, 2005) does not

solve the problem of external accountability, i.e. the risk of

exploitative capture of public resources by the partnership

as a whole (Jessop, 1998). Yet conventional institutional

forms of external accountability, such as an agency with a

funding program, reduce the self-organizing flexibility and

cross-partner communication that are the heart of colla-

boration (Jessop, 1998). A key research question then

concerns the possible institutional designs that can provide

accountability, stability and governability without killing

the flexibility necessary for continuous learning and

adaptation.

4.3. From learning to practice

Taylor and Short (2009), analyzing the Science and Ecosystem

Restoration Programs, raise a point overlooked by other

contributions focusing only on shared learning; the possibility

that the creation of a new language and new understandings

may not automatically translate into concrete actions that

solve the problems at hand.

Further analytical work is needed in the knowledge

generation literature to address the need to move develop-

ing shared understanding and to shared capability to apply

the understanding in a problem-solving mode (Taylor and

Short, 2009). Central here is the role of ‘‘bridge researchers’’,

individuals who understand how agencies work and switch

roles between research and implementation agencies,

cross-transferring new information and emerging under-

standings (Taylor and Short, 2009). Studies of collaborative

governance in the Everglades confirm the critical role of

such ‘‘super-agents’’, knowledge-brokers who serve in

multiple leadership roles of participating organizations,

inhabiting multiple knowledge spaces and helping bring

them together (Dengler, 2007). Importantly, such ‘‘leaders’’

in the Everglades had personal connections and the capacity

to communicate the knowledge generated in the process to

powerful state and national politicians ultimately respon-

sible for the funding and authorization of collaborative

plans (Dengler, 2007).

e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 2 ( 2 0 0 9 ) 6 3 1 – 6 4 3 639

5. Fundamental limitations and challenges

5.1. Evaluating governance

The inability to demonstrate tangible results for the $3 billion

invested in the program was a central reason for CALFED’s

dissolution (LHC, 2005). But did CALFED fail or succeed in

ways that were not immediately obvious? Some contributors

in this Issue refer to the deterioration of environmental

conditions and the lack of progress is supply reliability and

seismic vulnerability as signs of failure. Others point to

innovative agreements, unprecedented dialogue and new

understandings as signs of success. Many would agree with

Bobker’s (2009) argument that CALFED succeeded in trans-

forming cultures, processes, languages and understandings

but failed in delivering the desired management and policy

outcomes.

Bobker (2009) makes a distinction between process as the

means, and substantive goals as the ends of a policy. However,

such a distinction between procedural means and substantive

ends is hard to maintain. Governance emerges precisely to

negotiate disagreements over incommensurable ends and to

deal with the immense complexity of the system that makes

hard any assessment of whether it gets better and why. For

example, it is still contentious whether the decline in Delta

smelt is indeed an indication of the environmental health of

the system. Even less certainty exists over the causes of its

decline; pumps and water exports seem to play an important

role but noise from natural, long-term variations and

synergies with other factors such as pollution, invasives or

in-Delta water uses are hard to establish. Even more difficult

would be to weigh – uncertain or temporary – improvements in

some goals (e.g. salmon populations) against – uncertain or

temporary – deterioration in others (e.g. water exports).

Process in this sense has a value in and of itself. Continuously

negotiating, learning and adapting goals and metrics in the

face of disagreement and changing conditions is a sign of

success from governance’s own logic (Jessop, 1998). Hence,

Connick and Innes (2003) propose a new set of evaluation

criteria for collaborative processes such as production of

‘‘high-quality agreements’’, ‘‘innovation’’ or ‘‘learning’’.

Evaluating governance in its own terms however is also

unsatisfactory: Bobker (2009) points out that procedural

benefits should ultimately produce substantive improvements

to the managed system in a politically acceptable timeframe.

External accountability of the partnership needs a set of

external goals, other than the presence of the partnership

itself, upon which its performance can be evaluated.

The question of how to evaluate collaborative governance

and adaptive experiments cannot be resolved by theory or in

the abstract. It is through a pragmatic mix of both procedural

and substantive evaluations (Bryson et al., 2006), internal and

external to the process, that the accountability and legitimacy

of a collaborative process can be – always provisionally –

maintained. The research task here is to collect the experience

with evaluation practice and problems from different colla-

borative governance and AM experiments around the world,

and draw lessons about which evaluative arrangements work

best, at what scales and under what geographical and socio-

political conditions.

5.2. The dark side of collaboration

Exposing the ‘‘dark side of networks’’ (McGuire, 2006), Shilling

et al. (2009) argue that CALFED process fortified the privileged

access to California’s water policy by the ‘‘iron triangle’’ of

agencies, urban users and irrigated agriculture, together with

the emergent power of mainstream environmental groups. In

the process, weaker actors were marginalized, including the

broadly defined Environmental Justice (EJ) community repre-

senting low-income groups, small farmers, native commu-

nities, people of color, and more radical environmental

interests. Authorities treated EJ issues uncomfortably, dele-

gating them to a specialized workgroup with an unclear

mandate, little influence on decisions and isolated from the

rest of CALFED (Shilling et al., 2009). Shilling et al. highlight the

difference between collaborative governance CALFED-type

and ‘‘participatory governance’’ (Fung, 2006), defined as ‘‘the

active involvement of citizens in government decision-

making’’ (O’Leary et al., 2006, p. 7). Public overseeing

committees with open meetings, such as the BPAC and the

EJ working-group, did not provide real citizen input and

remained peripheral to the process (LHC, 2005; Shilling et al.,

2009).

Shilling et al.’s (2009) findings speak to a broader literature

critical of collaborative governance which argues that rather

than democratizing decision-making, collaborative arrange-

ments increase democratic deficit by strengthening those

who are able to exercise greater access and expertise in

relation to the new governance mechanisms (Lemos and

Agrawal, 2006; Swyngedouw, 2005). These critical works call

into question the legitimacy and representative account-

ability of the powerful NGOs that come to speak for citizens or

the environment in these processes (Swyngedouw, 2005).

They also show how the new inter-language that emerges

within collaborative processes is, first, partly conditioned by

the initial framing, mandate and participant composition of

the process, and second, privileges those actors who are

willing to accept it and ‘‘play by the rules’’, while margin-

alizing those with more radical views (Swyngedouw, 2005; de

Angelis, 2003). From this critical perspective, solutions such

as the EWA, which appear to collaboration scholars as

innovative and consensual (Lejano and Ingram, 2009; Innes

et al., 2007), are seen instead as an expansion and legitimiza-

tion of the dominant market logic and language to the

environmental realm, perpetuating past injustices (Gibler,

2005). With the EWA for example, environmental agencies are

forced to think in market terms and trade what previously

was a nonnegotiable regulatory limit for protecting endan-

gered species (Gibler, 2005). Worse, agribusinesses get paid by

taxpayers to give up water that they received subsidized from

the State (Taugher, 2009).

Others see such arguments as overblown. In their view,

first, collaborative governance is a pragmatic response to a

resource conflict, and it makes sense to involve primarily

those with power to end a stalemate (Fullerton, 2009). The

EWA, for example, even if unfair from a historical perspective,

was the only feasible solution to a stalemate where agricul-

tural interests refused to give up their water rights, environ-

mental agencies had problems limiting water pumping, and

courts offered unpredictable solutions devoid of nuance and

e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 2 ( 2 0 0 9 ) 6 3 1 – 6 4 3640

satisfactory to no one (Innes et al., 2007; Freeman and Farber,

2005). Second, the groups that fail to access collaborative

governance typically have also limited access to the arenas of

representative democracy too. Third, collaborative govern-

ance after all is not meant to substitute, only complement,

institutions of representative democracy, such as the legis-

lature (Fullerton, 2009).

This is, however, a rather simplistic view of the complex

interaction and articulation between representative democ-

racy and collaborative governance (Swyngedouw, 2005).

Governance processes, as evident in CALFED, leave much to

be desired in terms of transparency, accountability and public

oversight, and this partly has to do with their very nature that

rests on informal, ad hoc and closed-door interactions.

Governance is not a risk-free complement to conventional

governing institutions. First, governance takes time and

money out of normal governing from agencies with stripped

financial and human resources. Second, governments may

use governance processes to evade taking controversial

decisions (Hanemann and Dyckman, 2009) or to diffuse

political responsibility and accountability. Or in other

instances, they may tactically use it to rubber-stamp con-

troversial decisions, justify exclusion of uncooperative actors

or even scapegoat collaboration itself, offering a pretext for

return to authoritative modes of governing (Swyngedouw,

2005; de Angelis, 2003).

Beyond this fundamental debate, some argue in more

pragmatic terms that unless those with marginal, extreme

views (on both ‘‘sides’’ of the spectrum) are left out of the

negotiation there is the risk of reproducing sterile philoso-

phical debates, increasing frustration, and derailing action

(Innes et al., 2007). Yet this view of democracy is in tension

with the view of a discursive democracy held by Norgaard et al.

(2009) as one that puts fundamental philosophical and value

standpoints under deliberation. As Swyngedouw (2005)

argues, politics is precisely about debating fundamentals

and exposing conflict. It is in this sense that Norgaard et al.

(2009) argue for engaging deeper into the conflicts swirling

around the Bay-Delta with the fundamental tensions of

inclusive, deliberative processes rather than shying away

from them. The question is how to enhance the democratic

and inclusive character of processes like CALFED and how to

develop effective bridges with other forms of governing

(government, legislature, courts, etc).

5.3. Distribution, governing and governance

A broader notion of ‘justice’ includes the distribution of the

costs and benefits of environmental change between different

groups (Shilling et al., 2009). Hanemann and Dyckman (2009)

argue that negotiation over Bay-Delta ecosystem’s goods and

services is fundamentally a zero-sum game. The literature on

collaboration suggests that by bringing adversary parties

together, either win–win solutions can be devised, or that

adversaries will see the problem differently in ways that will

shift the focus from trading interests, to achieving shared

benefits (Hanemann and Dyckman, 2009; Fuller, 2009; Free-

man and Farber, 2005; Connick and Innes, 2003). Yet the

presence of a sufficiently large win–win space of shared

benefits cannot be assumed a priori. Hanemann and Dyckman

(2009) make a strong empirical-historical case that the Bay-

Delta involves a fundamental opposition of interests. Yet

other contributions describe smaller sub-processes initially

considered zero-sum in which mutually beneficial agreements

were reached (Fuller, 2009; Lejano and Ingram, 2009). A

remaining question is whether such win–win agreements

amounted to much more than ‘‘tinkering at the edges’’

(Fullerton, 2009, see also Owen, 2009; Brown and Kimmerer,

2009). Second, it seems that the win–win space for these

agreements may have been built by state subsidies, with the

tax-payers as potential losers unless the agreements delivered

collective benefits. It is reasonable to extend with the

arguments of Hanemann and Dyckman, and argue that

solving the problems of the Bay-Delta, like many other

environment-development conflicts around the world,

involves fundamental choices concerning what should a

desirable ecosystem look like, who gets to do what with their

land and water, who pays and who benefits. CALFED failed in

dealing with these core issues, as manifested in the failure to

establish a user-fee arrangement to finance the program (LHC,

2005). Hanemann and Dyckman (2009) conclude that it is

ultimately the responsibility of the State to make tough

distributive choices about the collective good. Collaborative

efforts may be symptomatic of policy-makers avoiding hard

decisions.

A criticism of this ‘‘return to State authority’’ thesis is that

mediation and collaboration in CALFED itself arose from

endless rounds of litigation sparked by decisions the State did

make. It is not clear how greater State ‘‘decision-making

capacity’’ can overcome such fragmentation of interests

unless greater raw power is exerted to suppress those whose

interests would suffer. Furthermore, the story of the SWRCB

(Hanemann and Dyckman, 2009) may be read not so much as a

lack of authority, but as its selective (mis)use: when the

SWRCB was ready to use its authority and set limitations to

water users in 1985, the Governor used his to maintain the

status quo.

Calls for greater state authority (e.g. LHC, 2005) underplay

that States are not infallible, even when they do have and

exercise decision-making capacity (the U.S. Federal Reserve

and the recent financial crisis are obvious examples). They

also ignore the broader political-economic context of free

market (neo-liberal) policies in the 1980s and 1990s that

deliberately weakened State capacities to govern public goods,

such as environmental protection. Beyond calls for a shift

from governance back to government and calls for authority

and leadership (LHC, 2005), what is needed is a strengthening

of the capacity, political responsibility, accountability and

representative character of conventional government, a

prerequisite for meaningful collaborative governance too.

Governance is not a substitute for a good, fair and effective

government.

5.4. Control and adaptation

Owen (2009) makes the provocative argument that even if it

endorsed an experimental approach, CALFED was rooted in

the belief that the Bay-Delta can be re-engineered and

controlled to the limit to satisfy competing needs. His work

points to an interesting contradiction within the logic of AM in

e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 2 ( 2 0 0 9 ) 6 3 1 – 6 4 3 641

that while AM recognizes the inherent uncontrollability of

complex socio-ecosystems, the objectives of experimentation

are ultimately to improve our capacity to understand and

better control the system. Owen instead espouses a precau-

tionary approach of limiting human intervention on ecosys-

tems given the limitations of our understanding. In the case of

the Bay-Delta this includes reducing water consumption and

withdrawal, though one could extend the argument to include,

more generally, reductions in the emission of pollutants or

controls to the urbanization of the Delta. Interestingly, Owen

does not base his argument on a romantic environmentalist

ideal of untouched ecosystems as being inherently better. His

argument is that reducing the intensity of consumptive uses of

the Bay-Delta is likely not only to improve environmental

conditions, but also to increase the stability and reliability of

consumptive patterns, although at a lower level.

An AM experiment that could have improved the ecological

conditions of the Delta may indeed have entailed reducing

diversions dramatically, reconverting much of the farmland in

the watershed to habitat and/or eliminating most toxic

discharges in the basin (Fullerton, 2009). No experimentation

was made with these most significant management ‘knobs’,

not only because of entrenched interests or risk aversion of

decision-makers; as Fullerton (2009) argues stopping water

flowing south is simply a politically infeasible option and

eventually a socially unpopular one given that most Califor-

nians are unlikely to trade their material standards for

environmental benefits. The AM literature overlooks such

real-life constraints to experimentation and does not enter-

tain the possibility that under multiple, incongruent goals and

constraints the spectrum of experimentation might be very

limited.

Furthermore, whereas the AM literature assumes that

collaboration and networking are good for experimentation,

CALFED shows that collaboration by its very nature tends to

sideline the more radical political options for the sake of

common ground. Innovation therefore is permitted within a

limited, potentially win–win space, while radical innovation a

priori excluded. Others too have noticed that whereas

collaborative governance may produce agreement over

techno-managerial solutions that promise to improve the

efficiency of resource use it is unlikely to promote more radical

options that involve restraints in human consumption (Lemos

and Agrawal, 2006). This begs the question concerning the

type of State-governance arrangements that may produce the

radical experiments and changes that might be necessary not

only in the Bay-Delta, but more generally with respect to

climate change and the other environmental crises that

economic growth and rising consumption are producing.

6. Conclusions

CALFED helps us see the benefits of collaborative, adaptive

governance and what it takes to achieve them, but it also helps

us see some limitations. Informality, self-organizing interac-

tion and sustained boundary work are some of the conditions

for success. Less clear is what sort of institutional designs can

create and maintain these conditions, while assuring that

agreements will be implemented in a publicly and politically

accountable way. The interactions and division of responsi-

bilities between new forms of governance and existing forms

of governing emerge as a key question; governance may be

capable of generating new and innovative ideas, but appears

ill-suited for dealing with core distribution issues that are at

the heart of water conflicts. The democratic deficit of

governance processes further undermines their legitimacy

in resolving distributive dilemmas. Governance is no sub-

stitute for governing and the need to reform and improve the

accountability, effectiveness and inclusiveness of conven-

tional government. Research needs to go beyond treating

collaborative and adaptive governance as distinct processes,

celebrating their successes or exposing their failures. The task

ahead is to rethink the boundaries of governing and govern-

ance and imagine and evaluate effective arrangements that

distribute functions appropriately optimizing results.

Furthermore, it might be wise to start thinking whether

slowing down the drivers of change might be more appro-

priate rather than attempting to live with optimized but

imperfect governance. Stakeholders in the Bay-Delta have

reached the painful realization that they cannot have it all.

Environmental conservation or restoration and further growth

appear fundamentally at odds. Climate change and resource

depletion suggest that current levels of consumption in the

developed world may be unsustainable in the longer term. A

reduction in consumption patterns might not only reduce

environmental pressures, but also vulnerability, securing the

reliability of infrastructures and consumption at a lower level

(Owen, 2009), sufficient to satisfy basic needs. Nonetheless,

radical changes reducing consumption and slowing growth

appear politically and socially impossible, if not utopian,

within present forms of representative government and

governance, alike. The question then concerns the type of

democratic institutions and political reforms necessary to

facilitate such radical, yet necessary, changes in California

and elsewhere.

Acknowledgements

This paper benefited from the comments of Boyd Fuller, Dave

Owen, Helen Ingram and Raul Lejano. We are grateful to the

authors, commentators, and reviewers for their strong

contributions to this Special Issue, and to everyone involved

for their patience and support.

r e f e r e n c e s

Agranoff, R., 2006. Inside collaborative networks: the lessons for public managers. Special issue. Public Administration Review 66, 56–65.

Bobker, G., 2009. The means do not justify the ends. A comment on CALFED. Environmental Science and Policy 12 (6), 726–728.

Brown, L., Kimmerer, W., et al., 2009. Managing water to protect fish: a review of California’s environmental water account, 2001–2005. Environmental Management 43 (2), 357–368.

Bryson, J.M., Crosby, B.C., Stone, M.M., 2006. The design and implementation of cross-scale collaborations: propositions from the literature. Public Administration Review 2006, 44–55.

e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 2 ( 2 0 0 9 ) 6 3 1 – 6 4 3642

CALFED Bay-Delta Program, 2000. Programmatic Record of Decision. CALFED Bay-Delta Program, Sacramento, CA.

Connick, S., 2003. The Use of Collaborative Processes in the Making of California Water Policy: The San Francisco Estuary Project, the CALFED Bay-Delta Program, and the Sacramento Area Water Forum. Environmental Science, Policy and Management. Ph.D. Dissertation. University of California, Berkeley, 496 pp.

Connick, S., Judith, I., 2003. Outcomes of collaborative water policy making: applying complexity thinking to evaluation. Journal of Environmental Planning and Management 46 (2), 177–197.

de Angelis, M., 2003. Neoliberal governance, reproduction and accumulation. The Commoner (Spring/Summer), 1–28.

Dengler, 2007. Spaces of power for action: governance of the everglades restudy process (1992–2000). Political Geography 26 (4), 423–454.

Dietz, T., Ostrom, E., Stern, P., 2003. The struggle to govern the commons. Science 302, 1907–1912.

Duane, T.P., 1997. Community participation in ecosystem management. Ecology Law Quarterly 24, 771.

Folke, C., Hahn, T., Olsson, P., Norberg, J., 2005. Adaptive governance of socio-ecological systems. Annual Review of Environment and Resources 30, 441–473.

Freeman, J., Farber, D.A., 2005. Modular environmental regulation. 54 Duke Law Journal 795.

Fuller, B., 2009. Surprising cooperation despite apparently irreconcilable differences: agricultural water use efficiency and CALFED. Environmental Science and Policy 12 (6), 663–673.

Fullerton, D., 2009. Tinkering at the edges. Environmental Science and Policy 12 (6), 733–736.

Fung, A., 2006. Varieties of participation in complex governance. Public Administration Review.

Gibler, J., 2005. Water for People and Place. Public Citizen, San Francisco.

Gottlieb, R., 1988. A Life of its own. The Politics and Power of Water. HBJ Publishers.

Gunderson and Light, 2006. Hanemann, M., Dyckman, C., 2009. The San Francisco Bay-

Delta: a failure of decision-making capacity. Environmental Science and Policy 12 (6), 710–725.

Healey, M.C., Dettinger, M.D., Norgaard, R.B. (Eds.), 2008. The State of Bay-Delta Science. CALFED Science Program, Sacramento, CA.

Heikkila, T., Gerlak, A.K., 2005. The formation of large-scale collaborative resource management institutions: Clarifying the roles of stakeholders, science, and institutions. Policy Studies Journal 33, 583–612.

Holling, C.S. (Ed.), 1978. Adaptive Environmental Assessment and Management. International Institute for Applied Systems Analysis, and Wiley, Toronto.

Hudgik, C.M., Arch, M.A., 2003. Evaluating the Effectiveness of Collaboration in Water Resources Planning in California: A Case Study of CALFED, IURD Working Paper Series, WP 2003-06.

Hundley Jr., N., 2001. The Great Thirst: Californians and Water, A History, revised ed. University of California Press, Berkeley.

Ingram, H., Fraser, L., 2006. Path dependency and adroit innovation: the case of California water. In: Repetto, R. (Ed.), Punctuated Equilibrium and the Dynamics of U.S. Environmental Policy.

Innes, J.E., Booher, D., 1999. Consensus building and complex adaptive systems: a framework for evaluating collaborative planning. Journal of the American Planning Association 65.

Innes, J., Connick, S., Kaplan, L., Booher, D.E., 2006. Collaborative Governance in the CALFED program: Adaptive

Policy Making for California Water, IURD Working Paper Series, WP 2006 01.

Innes, J.E., Connick, S., et al., 2007. Informality as a planning strategy. Journal of the American Planning Association 73 (2), 195–210.

Jessop, B., 1998. The rise of governance and the risks of failure. International Social Science Journal 155, 29–45.

Kallis, G., Butler, D., 2001. The EU water framework directive: measures and implications. Water Policy 3 (3), 125–142.

Kettl, D.F., 2006. Managing boundaries in American Administration: the collaboration imperative. Public Administration Review 10–19.

Lejano, R.P., Ingram, H., 2009. Collaborative networks and new ways of knowing. Environonmental Science and Policy 12 (6), 653–662.

Lemos, M.C., Agrawal, A., 2006. Environmental governarnace. Annual Review of Environment and Resources 31, 297–325.

Little Hoover Commission, 2005. Still Imperiled, Still Important: The Little Hoover Commission’s Review of the CALFED Bay- Delta Program. Little Hoover Commission Report #183. November 2005. from http://www.lhc.ca.gov/lhcdir/ report183.html.

Lurie, S., 2004. Interorganizational Dynamics in Large-scale Integrated Resources Management Networks: Insights from the CALFED Bay-Delta Program. Unpublished Ph.D. Thesis. University of Michigan.

McGuire, M., 2006. Collaborative public management: assessing what we know and how we know it. Public Administration Review 66, 33–42.

Mount, J., Twiss, R., 2005. ‘‘Subsidence, sea level rise, seismicity in the Sacramento-San Joaquin Delta.’’ San Francisco Estuary and Watershed Science 3 (1): Article 5, http:// repositories.cdlib.org/jmie/sfews/vol3/iss1/art5.

Norgaard, R.B., Kallis, G., Kiparsky, M., 2009. Collectively engaging complex socio-ecological systems: re-envisioning science, governance, and the California Delta. Environmental Science and Policy 12 (6), 644–652.

O’Leary, R., Catherine, G., Lisa, B.B., 2006. Introduction to the symposium on collaborative public management. Special issue. Public Administration Review 66, 6–9.

Ostrom, Elinor, 1990. Governing the Commons: The Evolution of Institutions for Collective Action. Cambridge U. Press, Cambridge.

Owen, D., 2009. Legal constraints, environmental variability, and the limits of innovative environmental governance. Environmental Science and Policy 12 (6), 684–693.

Reilly, W.K., 2009. Time for the Governor to Wade into the Delta. San Francisco Chronicle, San Francisco, CA.

Rieke, E.A., 1996. The Bay-Delta accord: a stride toward sustainability. University of Colorado Law Review 67, 341.

Shilling, F., London, J.K., Levanos, R., 2009. Marginalization by collaboration: Environmental Justice as a Third Party in and beyond CALFED. Environmental Science and Policy 12 (6), 694–709.

Swyngedouw, E., 2005. Governance innovation and the citizen: the Janus face of governance-beyond-the-state. Urban Studies 42 (11), 1991–2006.

Taugher, M., 2009. Pumping water and cash from Delta, Contra Costa Times. 5/23/09.

Taylor, K., Short, A., 2009. Integrating scientific knowledge into large-scale restoration programs—the CALFED Bay-Delta Program experience. Environmental Science and Policy 12 (6), 674–683.

Thompson, A.M., Perry, J.L., 2006. Collaboration processes: inside the black box. Special issue. Public Administration Review 66, 20–32.

e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 2 ( 2 0 0 9 ) 6 3 1 – 6 4 3 643

Giorgos Kallis completed a Marie Curie international fellowship at U.C. Berkeley (2005–2008). He holds a doctorate in Environmental Policy and Planning from the University of the Aegean. He is currently an ICREA fellow at the Institute for Environmental Science and Technology at the Autonomous University of Barce- lona. His research interests include participatory methods for water resources decision-making, environmental co-governance, ecological economic coevolution and interdisciplinary analysis of droughts.

Michael Kiparsky is a Ph.D. candidate in the Energy and Resources Group (ERG) at U.C. Berkeley, where he works on both technical and policy aspects of the science–policy interface, with a topical focus on climate change and water resources. His dissertation

work focuses on an analysis of risks to water supply from climate change and other factors. He holds an MS from U.C. Berkeley, and an AB in Biology from Brown University. His work has been supported by a National Science Foundation Graduate Fellowship, a CALFED Science Fellowship, an ACWA Water Law & Policy Scholarship and a Udall Environmental Fellowship.

Richard Norgaard is a professor of Energy and Resources at U.C. Berkeley, earned his Ph.D. in economics at the University of Chicago, and is among the founders of the field of ecological economics. His current research emphasizes how scientists from different disciplines collectively understand complex systems through deliberative processes and the implications of this mode of science for management, policy, and democracy.