Siting of liquefied gas terminals

Material Information

Siting of liquefied gas terminals a rhetorical analysis of stakeholder narratives
Nemoy, Yelena
Publication Date:
Physical Description:
ix, 210 leaves : ; 28 cm


Subjects / Keywords:
Liquefied natural gas shipping terminals -- Location ( lcsh )
Discourse analysis, Narrative ( lcsh )
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )


Includes bibliographical references (leaves 203-210).
General Note:
School of Public Affairs
Statement of Responsibility:
by Yelena Nemoy.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
122938640 ( OCLC )
LD1193.P86 2006m N45 ( lcc )

Full Text
Yelena Nemoy
B.A., Pace University, 1998
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Public Administration

This thesis for the Master of Public Administration
degree by
Y elena Nemoy
has been approved
Michelle Comstock

Nemoy, Yelena (MPA, Master of Public Administration)
Siting of Liquefied Natural Gas Terminals: A Rhetorical Analysis of
Stakeholder Narratives
Thesis directed by Professor Lloyd Burton
The work utilizes Emery Roes narrative analysis framework to explore the content
of stakeholder narratives in the citing of Liquefied Natural Gas (LNG) facilities.
With the U.S. consumption of natural gas on the rise, the need for importation of
natural gas has increased. Liquefaction makes it possible to import large quantities of
natural gas by cooling and pressurizing it. LNG can then be transported by seafaring
vessels and regassified at the destination LNG terminals. One of the major
controversies in the siting of LNG terminals is caused by the conflict between the
need for gas as an affordable, relatively clean energy source and the public fears
about the safety of the LNG technology and threats LNG terminals can present to
populated areas and the environment. This work explores this controversy by
analyzing the content of narratives told by industry groups and community
opposition groups in order to uncover the prominent stories emerging on both sides
of the LNG siting issue, in hopes that such analysis can help energy policy makers
understand the nature of the fears about LNG technology that drive continued
resistance to LNG terminal siting in local communities. To achieve this goal, the
contents of eight websites created by stakeholders (FERC LNG website, three
websites created by sponsors of LNG terminals, and four websites created by groups

opposing these proposals) were analyzed using rhetorical criticism methodology.
The paper presents results of the analysis and draws conclusions about what
constitutes a successful narrative and the usefulness of qualitative approaches in the
field of policy analysis.
This abstract accurately represents the
recommend its publication.

My thanks to my advisor, Lloyd Burton, for his contribution and support to my work.
I also wish to thank the members of my committee, George Busenberg and Michelle
Comstock, for their valuable participation and insights.

1. INTRODUCTION......................................... 1
2. LIQUEFIED NATURAL GAS: OVERVIEW........................5
Natural Gas: U.S. Energy Needs and
Advantages of Liquefaction.........................5
LNG in the U.S.: Current and Proposed Facilities .11
Laws, Regulations, and Regulatory Agencies........15
Production, Transportation, and Storage...........19
LNG Safety .......................................25
FRAMEWORK........................................... 39
Emery Roe and Narrative Policy Analysis...........41
Dvora Yanow and Interpretative Policy Analysis ...49
Rhetorical Criticism..............................51
Ideas in the Public Sphere .......................53
4. METHODOLOGY...........................................58
AND AN EMERGING METANARRATIVE.........................65
FERC Story: Informative-Authoritative Narrative...66

Industry Story: The Sales Pitch
Weavers Cove.....................................80
Broadwater Energy.................................91
Dominion Cove Point..............................101
Opposition Story: Nonstories, Countemarratives
And a Call to Action................................Ill
Opposition to Weavers Cove:
Coalition for the Responsible Siting
Of LNG Facilities and Green Futures..............112
Opposition to Broadwater: Anti-Broadwater
Coalition and Save the Sound.....................135
Conservation Law Foundation:
An Emerging Metanarrative?..........................150
6. RECENT DEVELOPMENTS.....................................156
Federal Energy Regulatory Commission................156
Broadwater Energy...................................158
Weavers Cove.......................................163
7. DISCUSSION..............................................169
Introduction: Narrative Types.......................169
Power and Asymmetrical Narratives...................173
Policy and the Meaning of Narratives................181
A Note on NIMBY.....................................193
Why Do Narrative Policy Analysis?...................200


1.1 Narrative types and project status..................................170

The problem area to be investigated in this work is the controversy
surrounding the siting of Liquefied Natural Gas (LNG) terminals in the populated
areas and offshore coastal waters of the United States. While the policy issues
surrounding the siting of LNG terminals are complex and involve questions of public
safety, national security, economic policy, environmental safety, and federalism
(namely, states rights to manage development in their coastal waters and onshore
areas), this paper will concentrate on the discursive aspects of LNG policy, namely
the debate surrounding several LNG terminals that are currently in development.
One of the major controversies in siting of LNG terminals is caused by the conflict
between the need for affordable, clean energy source (and the increased dependence
of the United States on natural gas for its energy needs) and the public fears about
the safety of the relatively new LNG technology and concerns about the threats LNG
terminals present to populated areas and marine environments. This work will
explore this controversy through a qualitative analysis of stakeholder narratives
about LNG technology and the rhetoric used to shape them. The goal is to
uncover the prominent stories emerging on both sides of the LNG siting issue -

stories told by industry groups that propose new terminals, as well as stories told by
the opposition groups.
It is the hope of this author that understanding the content of these stories
could help policy makers and policy analysts to understand the nature of the fears
that persist about this technology and drive the continued resistance to LNG terminal
siting in local communities, as well as provide stakeholders with examples of
narratives that successfully (or unsuccessfully) use rhetoric to tell their stories and
advance their goals. It is this authors belief that analysts and policy makers need to
explore not only the scientific and economic data that help determine policy choices,
but also the ways in which affected communities make meaning of policy issues,
because of increased emphasis that many agencies place on public participation in
the policy-making process. With the growing availability of information that drives
policy choices on agency websites and from mass media sources, the public becomes
more sophisticated in its understanding of the policy process, while community
grassroots organizations become more politically savvy and knowledgeable about
ways in which they can affect policy choices and disseminate their message about
policy issues to the public at large. Understanding the ways in which these
communities and individuals construct their opinions about desirable and undesirable
policy choices (opinions which determine public support or opposition to specific
policies) can enable policy-makers to form partnerships with communities in seeking
out policy solutions. In this way, understanding the content and the nature of public

discourse about policy choices could help policy makers move from what is often an
adversarial stand-off with communities (especially when making siting decisions
about hazardous or otherwise undesirable facilities, as is the case with the subject of
this work) to a more participatory model of making policy decisions.
This paper will use qualitative data from the electronic media sources to
explore the ways in which industry stakeholders use information and language to
promote their projects and convince the public of the safety of LNG, as well as to
explore the ways in which affected communities make meaning of industry claims
about LNG technology, often expressing persistent fears of it, despite evidence of
satisfactory safety record of LNG technology. The paper will start by providing a
brief overview of the natural gas markets and LNG, including the design and safety
features of LNG vessels, storage facilities, and terminals. The following two sections
will present the theoretical and methodological framework behind this work and
explain just what is meant by the terms narrative and story, as well as provide
justification for the choice of data. The next section will concentrate on the analysis
of narratives of stakeholders affected by LNG terminal siting on both sides of the
issue in an attempt to identify and describe narrative themes and rhetorical devices
that are used by these.
Finally, the last sections will provide some overarching thoughts about the
role of power and politics in policy narrative construction, some ideas about what

analysis of narratives could offer to public officials and opposition groups, and an
update on recent key events surrounding terminal proposals discussed in this work.

This section will provide an overview of Liquefied Natural Gas (LNG)-
related issues in an attempt to establish the reasons behind the growing importance of
natural gas and natural gas importation to the U.S. economy and policy, the growing
industry interest in LNG technology, and the persistent concerns about the safety and
security of LNG vessels. First, this section will provide an overview of the U.S.
energy needs and the advantages of liquefaction in light of the growing need for
importation of natural gas from sources outside the U.S. Second, the section will
present the current state of LNG industry in the U.S., including the locations and
capacity of current facilities. Third, a brief overview of laws and regulatory agencies
involved in regulation of LNG facilities siting and operations will be discussed.
Fourth, the author will take a look at production, storage, and transportation of LNG,
including key safety issues involved in siting of LNG facilities.
Natural Gas: US Energy Needs and Advantages
of Liquefaction
In the United States, natural gas constitutes almost one quarter of all energy
used. During the last ten years, the U.S. produced between 85 and 90 percent of the
natural gas it consumed, with remaining balance imported via pipeline from Canada;
however, the production of natural gas in North America has been steadily declining.

Despite declines in production, the U.S. consumption of natural gas has been
on the rise, due to increased demand for natural gas in the utilities, industrial,
transportation, and residential sectors. Residential sector has seen an increase in
popularity of fuel-efficient gas appliances that have lower emissions. Similarly, the
industrial and transportation sectors are beginning to develop natural gas technology
as a cleaner alternative to other fossil fuels (natural gas consists of methane, CH4,
which is the simplest hydrocarbon compound). Utilities have been using generators
fired by natural gas to produce electricity, resulting in lower emissions and costs and
in higher fuel efficiency.1 When used as fuel, natural gas produces near zero particle
emissions and a reduction in carbon dioxide (C02) emissions by almost 70 percent
compared to coal. When used for power generation, natural gas results in near zero
sulfur dioxide (S02) emissions and a significant reduction in C02 emissions. Natural
gas evaporates into the air without leaving a residue and does not mix with water or
soil, meaning that it will not cause environmental damage if released into the air,
water, or soil. However, methane is a greenhouse gas and the use of natural gas
might have an effect on the global climate change.2
1 The U.S. Department of Energy. Office of Fossil Energy. 2005. Liquefied Natural
Gas: Understanding the Basic Facts. Washington, DC. (accessed January 10, 2006), 2.
2 Michot Foss, M. Center for Energy Economics. The University of Texas at Austin.
2003. LNG Safety and Security. Texas,
(accessed February 6,2006), 54.

Annual natural gas consumption in the US is projected to rise from 22.1
trillion cubic feet (Tcf) in 2004 to 30.7 Tcf in 2005. These projections, combined
with the decline in North American production of natural gas, give rise to fears that
resulting price volatility in the energy market and higher energy prices will have a
negative impact on the energy industry, consumers, and the US economy as a
Imports of natural gas have become an important solution to the decline in
domestic production and the spike in the US natural gas prices. Imports of natural
gas are expected to account for 19 percent (4.9 Tcf) of US consumption by 2010 and
for 28 percent (8.7 Tcf) by 2025. At the same time, worldwide consumption of
natural gas is expected to increase by about 2.2 percent annually and by 70 percent
by 2025, raising worldwide consumption of natural gas from 90 Tcf in 2001 to 151
Tcf in 2025. The global resources of natural gas are sufficiently vast to satisfy
projected demand and are estimated at more than 15,000 Tcf, including 6,079 Tcf of
immediately recoverable gas (about 60 times the current annual worldwide
consumption). However, the worlds supply of natural gas is distributed in a way
that is inconsistent with the demands of the market. Thus, the majority of natural gas
resources (58.4) percent are located in Russia, Iran, and Qatar, countries that account
only for 19.4 percent of worldwide consumption of natural gas. These countries
3 The U.S. Department of Energy, Liquefied Natural Gas, 2-3.

emerge as potential exporters of natural gas; however, their resources are stranded,
that is, located far from the markets with high demand for natural gas.4 The solution
to delivering natural gas to markets too far removed geographically from exporters to
be serviced by pipelines is liquefaction.
Liquefaction is a process of cooling natural gas to the temperature of -260F
and pressurizing it, resulting in a transformation of natural gas into a liquid, called
Liquefied Natural Gas (LNG). This results in a reduction in volume by a factor of
610 (1/600111 of original volume), which makes storing and transportation of LNG
across oceans possible. In addition to enabling delivery of natural gas to high
demand markets, liquefaction also makes it possible to store gas in proximity to the
market for future use during high consumption periods in areas where gas use
fluctuation is seasonal and pipeline capacity from supply sources is expensive (e.g.
areas such as New England, where cold snaps result in increased consumption during
the winter) or in areas where underground storage facilities are not present or cannot
be built due to geologic conditions.5
While attempts to liquefy natural gas have been ongoing since the 1900s, the
first commercial LNG plant was not built until 1941 (East Ohio Gas in Cleveland,
4 Ibid., 2-3.
5 Gaul, D. The U.S. Department of Energy. Energy Information Administration.
2004. U.S. LNG Markets and Uses: June 2004 update. Washington, DC. (accessed January 2006), 1, 19.

Ohio). The first LNG tanker ship, The Methane Pioneer, sailed from Lake Charles,
Louisiana, to Canvey Island, United Kingdom in 1959, demonstrating the possibility
of LNG transportation over ocean and marking the beginning of international trade
in LNG. The US entered the market in 1969, exporting LNG from the first marine
terminal in Kenai, Alaska, to Japan.6 During the same time, the US began importing
LNG from Algeria. However, the growth of domestic natural gas production resulted
in the gas bubble of the 1980s and the importation of LNG was virtually
abandoned, until the growth of domestic consumption and the decline in production
led to a renewed interest in LNG.
Currently, LNG markets can be divided into two geographic regions: 1) the
Atlantic Basin, covering Europe, northern and western Africa, and the Eastern and
Gulf coasts of the US and 2) the Asia/Pacific Basin, covering South Asia, India,
Russia, and Alaska. Thirteen countries import LNG, with three countries in the
Asia/Pacific Basin (Japan, South Korea, and Taiwan) accounting for 67 percent of
global imports. While the countries in the Atlantic Basin account for only 33 percent
of global imports, the imports to these countries are expected to grow as they build
new terminals and storage facilities. On the export side, Asia/Pacific Basin countries
produced almost half of the worldwide LNG exports, while Atlantic Basin producers
6 Gaul, 2.

accounted for 32 percent of LNG exports in 2003 and were expected to increase their
capacity by 2007.7
The global LNG business makes up a value chain with four components:
Exploration and Production, Liquefaction, Shipping, and Storage and
Regassification. In order to attract investors, an LNG project must deliver gas into a
pipeline at the price of a unit volume that is at least equal the combined costs of
producing, liquefying, transporting, storing, and revaporizing the gas, plus the costs
of the capital needed to build necessary infrastructure.8 The liquefaction plant
accounts for the majority of the estimated $7-10 billion total costs of the LNG value
chain, with production, transportation, and regasification making up the balance.
Recent improvements to LNG technology resulted in significant drops in costs across
the value chain, including drops in capital costs for liquefaction plans from $600
million per ton of capacity in the 1980s to $200 million per ton of capacity in 2001,
as well as a drop in construction costs of LNG tankers from $280 million per ship in
1995 to $150 million per ship in 2004. While specific costs vary depending on
geographic location, in the United States the cost of building an onshore terminal is
7 The U.S. Department of Energy, Liquefied Natural Gas, 3-5.
8 Ibid., 8

projected to be over $400 million, with the cost of building an offshore terminal
expected to be significantly higher.9
LNG in the U.S.: Current and Proposed Facilities
The United States imported 506.5 billion cubic feet (Bcf) of LNG in 2003
and 652 Bcf in 2004, a 29 percent increase. Currently, there are five LNG receiving
and regasification facilities on the Atlantic and Gulf Coasts, with combined annual
peak capacity of 1.3 Tcf.10 Trinidad and Tobago accounted for about 75 percent of
US LNG imports in 2003, with Nigeria, Qatar, Oman, and Malaysia supplying the
remaining LNG.11
Currently, there are four major LNG facilities in the United States. The Lake
Charles facility in Louisiana, owned by Southern Union, received 238 Bcf of LNG in
2003, the largest volume of any U.S. terminal. The facility is located 48 miles from
the Gulf of Mexico on the Calcasieu River, in the vicinity of seven interstate
pipelines that supply the majority of major U.S. markets. This facility has one berth
that can receive one vessel of 900 feet long and 140 feet wide at a time, four
unloading arms, several vaporization trains (complete sets of equipment needed to
regasify LNG), and three storage tanks, each with 2.1 Bcf capacity. Southern Union
plans to add a second berth, enabling the terminal to receive two LNG tankers at a
9 Ibid.
10 Ibid., 6.
11 Gaul, 4-5.

time. Dominions Cove Point, the second largest US facility, is located on the
Chesapeake Bay in Maryland and has a storage capacity of 7.8 Bcf, ability to receive
two tankers at a time, and ten vaporization trains that regasify LNG at peak capacity
of 1 Bcf/day (same amount as the Lake Charles facility). Cove Point has access to
three interstate pipelines that deliver natural gas to Middle Atlantic States. The
Distrigas facility located in Everett, Massachusetts (near Boston), owned and
operated by Tractebel LNG North America, is the longest continually operating
facility in the United States with storage capacity of 3.5 Bcf. In 2003, it received 158
Bcf of LNG from Trinidad and Tobago. Almost 20 percent of New Englands natural
gas comes from this facility; it is also the only facility to distribute gas by trucks
directly to customers. The smallest U.S. facility, Elba Island facility in Georgia, is
owned and operated by El Paso Corporation and has five vaporization trains and four
terminals with capacity to receive 44 Bcf per year.12 Finally, Gulf Gateway Energy
Bridge, the first offshore facility located in the Gulf of Mexico, 116 miles off the
Louisiana coast, and owned by Excelerate Energy, began operations in March
12 Ibid., 5-6.
13 The U.S. Department of Energy, Liquefied Natural Gas, 6.

Currently, LNG imports represent 2.7 percent of the U.S. consumption of
natural gas,14 yet the projected demand for LNG is expected to grow to 6.4 Tcf per
year by 2025,15 resulting in calls for increasing the number of LNG facilities in the
United States. Most of the currently proposed facilities are located in the U.S. and
Mexican West Coast, the Gulf of Mexico, the Bahamas, and the U.S. and Canadian
East Coast. North American West and East coasts are attractive because locating
facilities in these areas will serve markets that are currently accessible only by long-
haul pipelines. The Gulf region is attractive because existing pipeline infrastructure
would allow investors to avoid infrastructure costs. Proposals in the Gulf region
tend to be larger and more expensive and are proposed by large U.S. and foreign
energy companies, such as ExxonMobil, Dutch/Shell, ChevronTexaco, Occidental
Petroleum, and ConocoPhillips, while the projects in the Northeast and on the West
Coast involve smaller terminals that are expected to deliver LNG to markets that are
currently experiencing higher gas prices than areas in the Gulf region.16
Overall, 42 proposed projects along the coasts of U.S., Mexico, Bahamas,
and Canada are currently in various stages of development and approval, with the
majority located in the U.S. Twelve proposals have been approved by the Federal
14 Gaul, 6.
15 The U.S. Department of Energy, Liquefied Natural Gas, 6.
16 Gaul, 7,11.

Energy Regulatory Commission (FERC): Lake Charles, LA (Southern Union),
Hackberry, LA (Sempra Energy), Freeport, TX (Cheniere/Freeport LNG Dev.),
Sabine, LA (Cheniere LNG), Elba Island, GA (El Paso Southern LNG), Fall River,
MA (Weavers Cove Energy/Hess LNG), Sabine, TX (Golden Pass ExxonMobil),
two facilities in Bahamas (AES Ocean Express and Calypso Tractebel), and three
facilities in Corpus Christi, TX (Cheniere LNG, Vista Del Sol ExxonMobil, and
Ingleside Energy Occidental Energy Ventures). Awaiting FERCs approval are
new or expansion facilities in Long Beach, CA (Mitsubishi/ConocoPhillips Sound
Energy Solutions), Logan Township, NJ (Crown Landing LNG BP), Bahamas
(Seafarer El Paso/FPL), Port Arthur, TX (Sempra), Cove point, MD (Dominion),
LI Sound, NY (Broadwater Energy -TransCanada/Shell), Pascagoula, MS (Gulf
LNG Energy LLC), Bradwood, OR (Northern Star LNG), Pascagoula, MS (Casotte
Landing ChevronTexaco), Cameron, LA (Creole Trail LNG Cheniere LNG),
Port Lavaca, TX (Calhoun LNG Gulf Coast LNG Partners), Freeport, TX
(Cheniere/Freeport LNG), Sabine, LA (Cheniere/Freeport LNG), Hackberry, LA
(Cameron LNG Sempra Energy), Pleasant Point, Me (Quoddy Bay LLC), and
Robbinston, ME (Downeast LNG -Kestrel Energy).17
17 Federal Energy Regulatory Commission. 2006. Existing and Proposed North
American LNG Terminals. Washington, DC.
(accessed 10 January 2006).

Laws, Regulations, and Regulatory Agencies.
In the United States, several agencies have roles in authorization and
regulation of LNG facilities, creating a complex regulatory environment comprised
of a multitude of state and local agencies operating under several statutes.
Complicating the regulatory picture is the separation of LNG facilities, for regulatory
purposes, into the onshore and the offshore types; the Federal Energy Regulatory
Commission (FERC), together with the U.S. Department of Transportation and the
U.S. Coast Guard are responsible for authorizing onshore facilities, while U.S. Coast
Guard and the Maritime Administration are responsible for authorizing offshore
facilities. In addition, the U.S. Department of Energy, the Minerals Management
Service, the Fish and Wildlife Service, and the Army Corps of Engineers have a say
in the approval of new LNG facilities. Finally, offshore facilities must also be
approved by bordering coastal states and, even after receiving federal agency
approval, must comply with air and water standards established by the
Environmental Protection Agency and state environmental agencies.18
Section 3 of the Natural Gas Act of 200519 establishes FERC as the lead
agency responsible for authorization of new onshore LNG facilities, while Section 7
of the Act empowers FERC to authorize construction of interstate pipelines that
18 Gaul, 9.
19 Natural Gas Act. 15 U.S.C § 3 (1938).

carry gas from LNG facilities. In addition, under the National Environmental Policy
Act (NEPA),20 FERC is required to prepare environmental impact statements and
conduct safety and environmental reviews of existing LNG facilities. Under the
pipeline safety laws,21 the U.S. Department of Transportation, Office of Pipeline
Safety, is responsible for promulgating safety regulations for the transportation and
storage of LNG in interstate commerce or foreign commerce. Finally, the
Magnuson Fishery and Conservation Act,22 the Ports and Waterways Safety Act of
1972, as amended,23 and the Maritime Transportation Security Act of 200224 require
the U.S. Coast Guard to ensure the security of ports and grant it authority to review
security plans of proposed LNG facilities, as well as some siting authority in relation
to management of ship traffic in and around an LNG facility.25
20 National Environmental Policy Act of 1969. Pub. L. 91-190, 83 Stat. 852.42
2,49U.S.C. §601.
22 Magnuson-Stevens Fishery Conservation and Management Act. Pub. L. 94-265,90
Stat. 331.
23 Ports and Waterways Safety Act of 1972. Pub. L. 92-340, 86 Stat. 424. 33 U.S.C §
24 Maritime Transportation Security Act of 2002.46 U.S.C. § 701.
25 Gaul, 9.

Under the National Environmental Policy Act (NEPA) and the Deepwater
Port Act of 1974 (DWPA), as amended,26 the U.S. Coast Guard is granted lead
agency status in authorization of offshore LNG facilities outside of state waters and
is responsible for preparing of environmental impact statement. In addition, the
Coast Guard is responsible for the security of LNG ships and facilities in the coastal
waters of the United States.27 DWPA also gives the Coast Guard authority over LNG
pipelines outside of state waters; the pipelines come under the jurisdiction of FERC
above the high water mark when they come onshore.28 The Maritime
Administration (MARAD) within the U.S. Department of Transportation is
responsible for authorization of deepwater ports, in cooperation with the Coast
Guard, as established by DWPA, which was amended by the Maritime
Transportation Security Act to apply to offshore LNG facilities and pipelines.29
Finally, several statutes protect states rights with regard to authorization of
LNG facilities. NEPA allows states to act as cooperating agencies with FERC during
the review of proposed LNG facilities, which allows the states to contribute to the
26 Deepwater Port Act of 1974. Pub. L. 93-627,88 StaL 2126. 33 U.S.C § 1501.
27 Gaul, 9.
28 Federal Energy Regulatory Commission. LNG Laws and Regulations, Federal
Energy Regulatory Commission,
regs.asp (accessed February 4,2006).
29 Gaul, 9.

environmental review of proposed projects. In addition, states have the authority to
block construction of LNG facilities, even after authorization by federal agencies, by
denying permits under the Clean Water Act, the Clean Air Act, and the Coastal Zone
Management Act. Under Section 502 of the Clean Air Act,30 any person that intends
to operate a facility that is a source of air pollution is required to obtain a permit, as
outlined by the provisions of the Act; if a permit is denied by the responsible state
agency, the facility cannot be constructed. Under Section 401 of the Clean Water
Act,31 any person that intends to operate a facility that may result in a discharge into
navigable waters must obtain a certificate of compliance with the state water quality
standard from the responsible state agency; if the certificate is denied, the facility
cannot be constructed. Under Section 404 of the Clean Water Act,32 the U.S. Army
Corps of Engineers must issue a permit for discharge of dredged and fill material;
in order to obtain the Corps permit, an applicant must first obtain certification
outlined in section 401 of the Act. Finally, under Section 307(c) of the Coastal Zone
Management Act,33 anyone proposing to construct an LNG facility must certify that
the proposed activity in a designated coastal zone complies with the enforceable
30 Clean Air Act 42 U.S.C. 7401 § 501
31 Clean water Act of 1977. Pub. L. 95-217,91 Stat 1566. 33 U.S.C. 1251 § 401.
32 Clean water Act of 1977. Pub. L. 95-217,91 Stat 1566. 33 U.S.C. 1251 § 404.
33 Coastal Zone Management Act of 1972. Pub. L. 89-454, title III, as added Pub. L.
92-583,86 Stat. 1280. 16. U.S.C. 1451 § 307(c).

policies of the affected states coastal zone management program. If the state
refuses to agree with the certification, no Federal permit and/or authorization,
including FERC and U.S. Army Corps of Engineers authorization, may be granted.34
In addition, the Code of Federal Regulations, in the section entitled
Liquefied Natural Gas Facilities: Federal Safety Standards35 establishes safety
standards for LNG facilities that transport natural gas by pipeline and, in the section
entitled Waterfront Facilities Handling Liquefied Natural Gas and Liquefied
Hazardous Gas36 regulates marine transfer areas for LNG of new and proposed
waterfront facilities handling LNG.37
Production, Transportation, and Storage
While natural gas consists of methane (CH4), the simplest hydrocarbon
compound, LNG is 85 to 95 percent methane, along with ethane, propane, and
butane and nitrogen, which together comprise five to fifteen percent. LNG is
produced in a complex multi-step process in an LNG production plant consisting of
several components. First, gas arriving via pipeline from the gas field must be clean
34 Federal Energy Regulatory Commission, States Rights in Authorization of LNG
Facilities, Federal Energy Regulatory Commission, (accessed
February 4,2006).
3549C.F.R. § 193 (2005).
36 33 C.F.R. § 127 (2005).
37 Federal Energy Regulatory Commission, LNG Laws and Regulations.

and dry, which is achieved by stripping it of hydrocarbon liquids, hydrogen sulfide,
carbon dioxide, and dirt. Second, the gas is cooled, resulting in condensation of
water, which is necessary for dehydration and complete removal of water vapors. At
this stage, mercury must also be removed and the gas must be filtered before
liquefaction. During liquefaction, the gas is cooled using heat exchangers, in which
gas circulates through aluminum tube coils and is exposed to a compressed
hydrocarbon-nitrogen refrigerant, which cools the gas to -260F as it vaporizes. The
liquefied gas is then pumped into insulated storage tanks where it is stored at
atmospheric pressure. LNG can now be boiled at atmospheric pressure (the
temperature of LNG will remain constant), giving off LNG steam that must be
removed and that, at about 0.15 percent of volume per day, fuels the liquefaction
facility, regasification terminals, and LNG vessels. LNG can now be transferred to
the vessels using pumps that are built to withstand the cryogenic temperatures of
LNG. Currently, there are eighteen liquefaction plants worldwide, operating 71 trains
with an additional fourteen trains under construction.38
Safety concerns, particularly risks associated with spills, play a major role in
the design of LNG vessels and facilities. LNG vessels are double-hulled, about 1,000
feet long, contain up to five LNG tanks that can each hold several thousand cubic
38 The U.S. Department of Energy, Liquefied Natural Gas, 3,10-11.

feet of LNG, and require a minimum water depth of 40 feet39 There are two major
designs currently in use: the membrane design, which consists of multiple tanks with
linings made from 0.5mm nickel steel alloy designed to withstand extreme
temperatures incorporated into the hull of the ship, and the spherical or Moss
design that has round tanks mounted on supports on the hull of the ship that transfer
the stress of thermal expansion and contraction onto those supports.40 The cargo
tanks can be described as Thermos bottles: LNG is injected into the tanks where it
is stored under atmospheric pressure, with the tank insulation, up to two feet thick,
keeping the cargo cold. LNG vessels are designed with two safety systems: the
cargo-handling system designed to handle the low temperatures of the cargo and
the ship-handling systems designed to deal with the dangers associated with spills
and fires. The former include instruments that shut down the system if it starts
operating outside the specified parameters, while the latter include radar and
positioning systems, distress systems, automatic beacons, and fire and gas detection
Onshore receiving and regasification terminals occupy 25 to 40 acres and
consist typically of a docking structure, LNG storage tanks, and vapor-handling
39 Marks, M. et al. State of California Energy Commission. 2003. Liquefied Natural
Gas in California: History, Risks and Siting. Sacramento. (accessed January 10,2006), 3.
40 The U.S. Department of Energy, Liquefied Natural Gas,12.
41 Marks et al., 3.

systems. In addition, they require the use of infrastructure such as roads, electric
transmission lines, gas pipelines, and water lines. Docking facilities usually contain a
pier that is about 1,800 feet long and 30 feet wide with moorings that connect the
vessel to the jetty and off-loading structures that usually have peak flow rates of
about 12,000 cubic meters per hour and take 12-15 hours to unload a vessel. Jetties
have built-in safety features, designed to prevent release of LNG during unloading
due to tidal or wave forces, such as rapid ship-to-shore emergency shutdown (ESD)
systems and shut-off valves that will stop the flow of gas in 20 to 30 seconds.
Transfer pipes are made of stainless steel and are designed to withstand cryogenic
temperatures and rapid temperature drops.
Typically, LNG is stored on land before regasification. Storage tanks are
designed to prevent failure, which could result in releases of amounts of LNG of
more than 100,000 cubic meters and resulting vapor clouds. Currently, there are
three types of LNG storage tanks in use: single-containment tanks that are double-
walled, with interior tank made of nine percent nickel and the exterior tank made of
carbon steel; double-containment tanks that consist of a primary and secondary tank,
the latter a concrete wall located six meters or less from the primary tank and
containing cryogenic liquid that can limit the surface area and vaporization of an
LNG liquid pool; and full-containment tanks that consist of a nine-percent nickel
inner tanks and pre-stressed concrete outer tanks, which have concrete roofs lined
with carbon steel and are designed to withstand impact from missiles or flying

objects. All of these tanks have no side or bottom penetrations. That is, pipes that
carry LNG in and out of the tank are connected to the roof of the tank in order to
reduce the amount of potential spill in the event of a leak or rapture. In addition, all
tank designs include instruments to monitor temperature, density, and pressure of
LNG along the height of the tank in order to detect rollover conditions that occur
when LNG stratifies into layers of different densities within the tank and can result
in pressure increases that could cause damage to the tank. Tanks also contain
cameras that allow the operators to assess tank damage in case of an earthquake; fire,
pressure, temperature, and smoke detectors; and cameras that can identify spills.
Finally, tanks are contained within a walled system that would drain all spills into
basins of reinforced concrete and are surrounded by impoundments that are built to
contain at least 110 percent of the tanks volume if sudden failure occurs. These
impoundments limit the spread of a spill and reduce the surface area of the liquid
pool, therefore limiting the size of a potential vapor cloud.42
Since water provides a heat source and, therefore, speeds up vaporization,
water cannot be used in fighting LNG fires. Thus, LNG facilities include structures
that automatically deliver high-expansion foam to spill areas in order to disperse
LNG vapors upwards (since ignition sources are typically found on the ground) and
provide control over flames by covering the LNG pool surface to prevent heat
42 Marks, et. al, 3-6.

I transfer from the air to the liquid, thereby reducing the rate of vaporization and
limiting the rate of burning.43
i Finally, LNG facilities are designed so as to assure adequate distances
| between LNG storage tanks, the storage areas and the jettys, and the vaporization
| areas and other structures, as well as adequate exclusion zones (the area
| surrounding an LNG facility in which an operator legally controls all activities)
| which ensure that communities immediately outside the facility are not endangered
| in case of a facility fire or a release of a vapor cloud. There are two types of
exclusion zones: thermal-radiation protection (from fires) and flammable vapor-
i dispersion protection (from vapor clouds that could drift to an ignition source and
ignite); the distances for both are established using the National Fire Protection
Association (NFPA) standards for storage and handling of LNG and by computer
| models that take into account context-specific factors such as humidity, wind speed,
j average gas concentration in the air, weather, and terrain conditions. For instance,
the thermal exclusion zone around the Cove Point facility is 1,600 feet, while the
j vapor-dispersion protection zone around the same facility is 1,017 acres.44
43 Ibid.
44 Ibid.

LNG Safety
As a liquid, LNG is hazardous; as a gas, it is combustible. Therefore, it
presents several environmental, safety, and security risks. According to the
California Energy Commission, LNG is hazardous because of its cryogenic
temperatures, flammability, and dispersion properties. Because LNG is extremely
cold, prolonged contact with skin will result in freeze burns. In its gaseous form,
LNG is classified as an asphyxiant in case of human exposure.45
According to FERC, specific risks of LNG arise from its chemical properties.
As mentioned previously, LNG consists primarily (85 to 95 percent) of methane,
which is flammable if its concentration in the air by volume reaches five to fifteen
percent. At five percent concentration, LNG is at its lower flammability level (LFL)
or not sufficiently concentrated for ignition. At fifteen percent concentration in the
air LNG is at is upper flammability level (UFL); above this ratio it is too
concentrated for ignition. At its normal boiling point at the pressure of 1 atmosphere
and the temperature of -259F, LNG can evaporate rapidly, producing 620 to 630
standard cubic feet of natural gas for each cubic foot of liquid and forming a vapor
cloud with a gravity of 1.7. At this boiling point, LNG vapor is 1.5 times more dense
than air at 77 F, which means that it is negatively buoyant and the cold LNG
vapor resulting from a spill will float on the surface of water or ground until it warms
45 Marks et al, 2-3.

to the temperature of -162F and becomes less dense than air, at which point it will
rise and disperse into the atmosphere at the concentration less than its LFL.46
The downwind distance traveled by LNG vapor cloud will depend on volume
and rate of the spill, as well as on the weather conditions and on whether the cloud
meets an ignition source. FERC notes that an event of sufficient magnitude to
rupture an LNG cargo tank may also provide ignition sources. If the cloud is
ignited, the flame will bum back to the vapor source and the distance traveled by
the cloud will not be significant. Because LNG is not toxic, FERC does not
anticipate the probability of long-term environmental damage if ignition does not
occur. If ignition does occur it can result in one of the three types of fire: pool fire,
which can occur when LNG forms a liquid pool and some of the evaporating liquid
is ignited and travels back to the spill, igniting the vapor cloud that is forming above
the pool of LNG; jet fire, which can occur if LNG is released from storage tank,
forming a gas jet that mixes with air as it is released; and flash fire, which can occur
if portions of the vapor cloud are ignited and combust before the cloud reaches its
46 Federal Energy Regulatory Commission. 2004. Consequence Assessment Methods
for Incidents Involving Releases from Liquefied Natural Gas Carriers. Washington,
DC. (accessed January 2006), 2-3.
47 Ibid., 3-4.

Other risks associated with LNG are explosions, Rapid Phase Transition, and
previously mentioned cryogenic effects. Explosions are not known to occur in open
areas; however, methane can become over-pressurized if it is confined (e.g. in the
ship or in buildings). If a vapor cloud is trapped in an area such as a chemical
process plant, which is crowded with various structures and equipment, and is
ignited, an explosion may occur. Because LNG is not pure methane, but contains
heavier hydrocarbons, the danger of overpressures is increased by slower dispersion
and lower energy required for ignition and detonation. Rapid Phase Transition or
RPT is a phenomenon observed in LNG experiments, during which LNG transitions
from liquid to gas phase almost instantly, perhaps from superheating due to lack of
nucleation sites which help with the formation of bubbles and promote boiling,
resulting in overpressure or rapid dispersion or puff of LNG into the atmosphere.
RPT can result in steam explosions in industrial plants. Finally, LNG can also pose
cryogenic risks to LNG carriers, resulting in deck and tank fractures.48
Finally, some fear that LNG vessels and facilities can become terrorist targets,
thus presenting a security risk. In order to minimize the terrorist threat, the member
countries of the International Maritime Organization of the United Nations adopted
the Ship and Port Facility Security Code in 2003, which requires LNG vessels and
ports to assess vulnerabilities and to develop security plans. U.S. Maritime
48 Ibid., 4-5.

Transportation Security Act of 2002 requires all ports to assess LNG facilities and
vessels and produce security plans that must be federally approved.49
While, according to FERC, the industry has maintained a good safety record
over the past 40 years and 33,000 LNG vessel voyages, this record is not
unblemished. Over the years, a number of LNG-related incidents have occurred, the
most significant as recent as 2004. On January 19,2004, in Skikda, Algeria, a steam
boiler exploded in the liquefaction production plant operated by Sonatrach, a
government-owned company, resulting in 27 deaths and 56 injuries, as well as
damage to the areas surrounding the plant and the plant itself.50 In addition to the
incident an Algeria, there have been a number of other LNG-related accidents.
Thus, between 1944 and 2005, thirteen explosions and fires have occurred at LNG
1944, Cleveland Ohio: a tank at the peak-shaving plant failed, spilling LNG
into the street storm-sewer system and resulting in an explosion and fire that
killed 128 people. It was later determined that the steel alloy with low-nickel
content that was used to build the tank became brittle when coming in contact
with cryogenic LNG.
1964 and 1965, Arzew, Algeria: lightning struck the forward vent riser of the
LNG vessel Methane Progress during loading of LNG (in 1964) and while the
vessel was at sea (in 1965), resulting in the ignition of vapor that was being
vented through the venting system. In both cases, the fire was extinguished
without resulting in loss of life or significant damage.
49 California Energy Commission, Liquefied Natural Gas Safety, California Energy
Commission, (accessed 4 February 2006).
50 Federal Energy Regulatory Commission, LNG Safety Record, Federal Energy
Regulatory Commission,
(accessed 4 February 2006).

1969, Portland, Oregon: an accidental removal of blinds from a natural gas
pipeline resulted in the flow of gas into the tank that was still under
1972, Montreal East, Quebec, Canada: after completing the defrosting
operation, the valves on the nitrogen line were not closed, resulting in a back
flow of natural gas from the compressor into the nitrogen line. As a result, the
compressor became over-pressurized and natural gas leaked into the control
room, where operators were allowed to smoke, resulting in an explosion that
occurred when an operator tried to light a cigarette.
1973, Staten Island, New York: a fire started in an empty storage tank under
repair, resulting in an increase in pressure inside the tank that was so rapid as
to cause the concrete dome on the tank to lift and collapse into the tank, killing
37 construction workers inside.
1979, Cove Point, Maryland: a natural gas leak resulted in an explosion,
killing one employee and injuring another and resulting in $3 million in
1983, Bontang, Indonesia: A closed valve on a blowdown line cased an
extreme overpressurization of the heat exchanger, resulting in an explosion.
1987, Nevada Test Site, Mercury, Nevada: during tests involving spills of
LNG on the U.S. Department of Energy Test Site, an accidental ignition of an
LNG vapor cloud occurred, damaging the polyurethane pipe insulation.
2004, Skikda, Algeria: an explosion in a steam boiler triggered a vapor-cloud
explosion and fire, resulting in loss of life and property damage.
2004, Trinidad, Tobago: a gas turbine at Atlantic LNGs Train 3 facility
exploded, necessitating evacuation of workers from the facility.
2004, Ghislenghien, Belgium: a pipeline carrying gas from Belgium to France
exploded, resulting in 23 deaths. The cause of the incident is still unknown, but
it appears that a contractor might have accidentally damaged the pipe.
2005, District Heights, Maryland: a house explosion was connected to
molecular differences in the imported LNG that the utility switched to in 2003.
2005, Nigeria: a 28-inch underground pipeline carrying LNG exploded,
resulting in a fire that spread over 27 square kilometers.51
In addition to these accidents, eight incidents involving leaks and spills have
occurred between 1965 and 1979.
51 California Energy Commission, Liquefied Natural Gas Safety.

1965, Methane Princess Spill: LNG discharging arms were disconnected
before the lines were completely drained, resulting in a spill of LNG liquid
onto the deck and causing fracture of the deck plating.
1965, Jules Verne Spill: an overflowing cargo tank resulted in an LNG liquid
spill in Arzew, Algeria, causing fracture of the cover plating of the vessel and
the tank.
1979, La Spezia, Italy: A rollover, where two layers of LNG with different
heat and density form and mix, resulting in a release of vapor, occurred,
causing a discharge of about 2,000 tons of LNG vapor and damage to the tank.
1974, Massachusetts Barge Spill: As LNG was being loaded onto the barge, a
power failure occurred, resulting in the closure of the main liquid line valves
and the leak of 40 gallons of LNG that caused fracturing of the deck plates.
1977, Aquarius Spill: LNG overflowed through the vent mast serving the
cargo tank during filling of the tank.
1978, Das Island, United Arab Emirates: a bottom pipe connection of an
LNG double-walled tank failed, resulting in a heavier-than-air vapor cloud
forming on the outer wall of the tank. The cloud did not ignite.
1979, Mostafa Ben Bouliad Spill: A check valve in the piping system of the
vessel failed during unloading at Cove Point, resulting in a small leak and
minor fractures to the deck plating.
1979, PoUenger Spill. LNG leaked from a valve and damaged plating of a
tank during discharge of LNG at the Everett, Massachusetts terminal.52
Because of risks associated with LNG, terminal proposals must include a
detailed risk assessment and a risk management plan that must be presented to the
lead federal agency during the process of site selection. Sophisticated modeling must
be performed to assess potential dangers, with risk-assessment modeling typically
containing a four-step process that includes hazards assessment, hazard-probability
assessment, worst-case assessment, and worst-case probability assessment. Hazards
assessments identify risks associated with shipping and land-base activities;
probability assessments predict the likelihood of such events happening at the
52 Ibid.

proposed facility; worst-case assessments evaluate the possibility of loss of life,
injury and property damage; and worst-case probability assessments determine the
probability of such events. It has to be noted that significant potential risks are
justified as acceptable, because their likelihood is usually so small as to be
considered implausible (e.g. one in a million).3
Shipping-related events usually include collision, groundings, navigational
errors, and mechanical failures and are predicted to result in spills caused by
leakage from joints, errors during connection/disconnection of unloading arms, and
ruptures in the cargo systems. Land-based events that could lead to LNG spills
include equipment failure and site-specific events, such as earthquakes or terrorist
acts. The U.S. Department of Transportation determines design criteria to account
for earthquakes and other geologic conditions, requiring LNG facilities to be
constructed to withstand earthquake ground motion with a 1-in-10,000-year
exceedance without loss of structural or functional integrity. After the September
11,2001 terrorist attacks, the U.S. Coast Guard began to require LNG vessels to
provide a 96-hour notice of arrival in order to conduct terrorist risk assessment.
While some call for risk assessment of new LNG facilities to include the likelihood
3 Marks et. al, 13.

of terrorist attacks or sabotage, such assessments are considered to be unreliable due
to unpredictable and random nature of terrorist activities.54
Environmental impact statements relating to air emissions, cold-water
discharge, land use and other aspects are usually prepared for any proposed LNG
facility. Such assessments have yet to become major obstacles to construction.
This is probably due to the fact that environmental impacts of LNG importation are
usually considered to be minimal. LNG facilities do not flare natural gas, but capture
boil-off gas and use it to fuel vessels and facilities. In the case of LNG facilities, air
emissions usually come from diesel-fired generators; however, these generators
result in minimal emissions, as they are usually two megawatts or less in power and
are designed to operate only in cases of interrupted power from the electric grid.
However, per Coast Guard regulations, LNG vessels must generate electric power
while in port, resulting in emission of air pollution from diesel generators of docked
LNG ships. Similarly, tug boats that escort LNG vessels into ports are powered by
diesel engines and can sometimes be the largest source of air pollution from
proposed LNG facilities.
LNG facilities do not utilize large amounts of water, nor do they produce
large amounts of waste. However, cold-water discharges associated with the heat-
exchanger regasification systems could have a negative impact on aquatic
environments if the discharge plume were significantly colder than the ocean water
54 Ibid, 13-14.

into which the discharge was flowing. Another potential threat to marine
environment can come from ballast water, which is added to a cargo vessels ballast
tanks when it is empty of cargo, in order to maintain the stability of the vessel.
Ballast water is discharged when vessels take on cargo. Discharge of water taken
into the tank from one body of water into another body of water can introduce non-
native marine organisms into local water that could have detrimental effect on local
marine environment. Because LNG vessels come into U.S. ports loaded with cargo,
they do not carry ship-ballast water and do not discharge ship-ballast water from
other ports into U.S. waters. However, LNG tankers do take on ballast water from
U.S. ports after unloading their cargo. Therefore, there is some concern that they
may carry non-native species back to their home ports. In addition, Tilling and
dredging activities to accommodate the large tankers in coastal waters can have
significant impact on water quality, as well as on biological organisms. While
impacts on wildlife will depend on site characteristics, potential for negative effects
on habitat, wetlands, and wildlife is there. Finally, LNG facilities might introduce
noise, traffic, and visual impacts into the surrounding communities.55
As the overview above indicates, the LNG industry has had a good safety
record overall. The safety record of the LNG shipping industry over the past 60 years
has been exemplary, with about 150 vessels making 80,000 carrier voyages over 100
55 Ibid., 15-16.

million miles resulting in eight spills (but no fires) and two groundings (but no
spills).56 The safety of LNG transportation is attributed to the design of LNG vessels,
which, unlike oil and chemical tankers like the Exxon Valdez (the grounding of
which in 1989 resulted in a major oil spill in Alaska), are double-hulled.57 The safety
record of onshore facilities, however, is mixed, with several accidents over the years,
most serious of which, in Skikda, Algeria, is considered to be the worst
petrochemical plant fire in Algeria in 40 years.
That accident illustrates the kind of damage to life (27 workers killed and 74
injured out of 12,000 employed at the plant) and property (destroyed processing
plant and damaged marine birth) that can result from a LNG-related accident. As a
result of such accidents, public fears persist about the potential for large-scale,
catastrophic spills and fires involving LNG and resulting in significant damage to
communities and the environment. In the past few years, several studies have
attempted to predict and model the conditions under which such incidents can occur
and the potential consequences of such events. One of the most oft-cited studies is
the Sandia National Laboratories report, Guidance on Risk Analysis and Safety
56 Hightower, Mike eL al. Sandia National Laboratories. 2004. Guidance on Risk
Analysis and Safety Implications of a Large Liquefied Natural Gas (LNG) Spill Over
Water. Albuquerque, (accessed January
10, 2006), 26, 28.
57 Parfomak, Paul. Congressional Research Service Report for Congress. 2004.
Liquefied Natural Gas (LNG) Import Terminals: Siting, Safety, and Regulation.
RL32073.Washington, DC. (accessed January
10,2006), 6.

Implications of a Large Liquefied Natural Gas (LNG) Spill Over Water
(subsequently referred to as the Sandia Report), which takes a risk analysis
approach to identifying the risks of accidental and intentional LNG spills on water.
While the Sandia report identifies both the mechanism of an LNG spill on
water and the potential sequence of events during a breach of an LNG vessel, the
authors note that, due to the safety record of the LNG transportation industry, there is
little historical data on tanker breaches and spills, making modeling of accidental
breaches difficult Conducting finite element modeling of 90 degree collisions with
oil tankers similar in size and weight to an LNG ship, the authors estimate that, for
an inner hull to be breached on collision, impact velocity must exceed five six
knots and penetration must be about three meters deep. However, as vessels tend to
remain joined after colliding, a collision at velocity as high as ten knots is expected
to produce a hole of only about one square meter in size from which LNG could
spill. Overall, the potential for a breach of the inner cargo tank resulting in a spill is
estimated to be minimal, with the authors concluding that the safety procedures
currently adopted by the industry are sufficient for dealing with an accidental
collision or grounding of an LNG vessel. The authors expect a rupture of an inner
tank and subsequent spill of LNG to result in a pool fire or an ignition of a vapor
cloud, both resulting in thermal radiation, which could be highly hazardous to

people, but which would be confined to a relatively small area around the location of
the spill.58
Intentional LNG breaches and spills are also hard to estimate, due to the
conservative estimates used in experiments on LNG spills. The authors identify the
following intentional breach scenarios: ramming, which is not considered to be a
significant risk due to the high velocity necessary to penetrate both hulls and the
cargo tank of an LNG vessel; triggered explosion, which is considered to be
particularly hazardous in shallow waterways, harbor bottoms, and terminals where
explosives could be placed; insider takeover or hijacking, which is considered to be
a credible threat particularly when insider information could be used to override
offloading safety systems to release LNG onto the ship, water, or terminals; and
external terrorist actions, (U.S.S. Cole-type attacks), which are considered to be
credible threats and might result in breaches and spills of various sizes, depending on
the type of weapon used and the location of the attack. Intentional attacks can result
in a variety of events, such as moderate to severe cryogenic damage to the vessel,
pool fires and vapor cloud ignition (both likely to occur because an intentional attack
will most likely produce an ignition source), all ranging in severity depending on the
type of the attack, weather and other conditions, and measures taken to minimize the
damage. Intentional spills cany risks that have to be managed by developing risk-
management, planning and mitigation^ (such as reducing the potential for, and
58 Hightower et. al, 43-46.

consequences of a spill) and security procedures such as inspections for explosives
and surveillance of LNG ships.59
The authors of the Sandia Report conclude that, while release of LNG vapors
is unlikely, vapor clouds could travel a distance from 1,600 meters to 2,500 meters
and that the most significant impacts of an LNG fire to public safety and property
are likely to occur within 500 meters (due to thermal radiation heat from fires) to
1,600 meters of a spill. These areas in particular are likely to be exposed to major
injuries and significant structural damage, including damage to people and
infrastructure. Beyond 1,600 meters (for large spills) only minor damage to property
is expected.60
Finally, the Sandia Report identifies several risk management techniques that
can minimize the risks of both accidental and intentional LNG spills, breaking up the
type and severity of risks by zone. Zone 1 contains areas where LNG tankers transit
narrow harbors or channels, pass under major bridges or over tunnels, or come
within approximately 250 meters of people and major infrastructure elements. In
this zone, the risks and effects of an accidental or an intentional LNG spill can be
significant and severe, particularly the risks associated with thermal radiation
from an LNG fire, which can have severe negative impacts on public safety and
infrastructure. In Zone 1, the approach to risk management should be the most
59 Hightower et. al, 49-63.
60 Ibid., 14-25.

rigorous and should include strategies for dealing with vapor clouds and fires, such
as vessel security zones, waterway traffic management, and establishment of positive
control over vessels. Zone 2 contains areas where LNG is present in broader
channels or larger outer harbors and within 250 to 750 meters of people and critical
infrastructure and where the risk of thermal radiation impacts is less significant. In
ZLone 2, the approach to risk reduction can be less rigorous and should include
strategies for dealing with effects of vapor clouds and fires such as community
warning signals, education programs and ensuring that shelters are available.
Finally, Zone 3 contains areas more than 750 meters away from major infrastructure
and population centers, where the risk of thermal radiation is minimal. Here,
approach to risk reduction strategies can be less extensive, and concentrate on
emergency response measures that are focused on dealing with vapor cloud
dispersion.61 As previously mentioned, the thermal exclusion zone around the Cove
Point facility is 1,600 feet or 487.7 meters and, therefore, falls within Zone 2 that
is, the structures just outside the exclusion zone can still suffer some thermal
radiation damage.
61 Ibid., 22-34.

This work will apply qualitative methodologies to the policy problem of
siting LNG terminals. While the policy of siting LNG terminals is influenced by a
multitude of economic, environmental, national security, and safety considerations
and is administered by a variety of state and local agencies, the continued resistance
of the public to the siting of LNG terminals indicates that a dissatisfaction with siting
standards continues to exist and that the fear of LNG technology is persistent and
wide-spread despite the evidence that supports the safety claims of the industry.
Understanding what is said about siting and how it is said can help us understand the
roots of these fears. While social sciences such as sociology, social psychology, and
anthropology often concern themselves with describing the content of public
discourse on various social issues and the values and opinions that drive it, discourse
has received relatively little attention in the field of public policy. Because the goal
of this work is to explore the desires, fears, and needs expressed by the stakeholders
in LNG siting issues from the perspective of a public policy analyst, it finds itself in
a relatively uncharted territory. However, the growing popularity of qualitative
methodologies in other social sciences has not gone unnoticed in this field. Thus,

there are both theoretical frameworks and methodological tools available to an
analyst interested in utilizing this approach.
Over the last several decades, some academic disciplines, particularly social
sciencies, have undergone a shift from positivist methodologies to what became
commonly referred to as post-positivist approaches. Simply put, positivist
methodologies start with an assumption that there exists an objective answer (truth)
to every scientific question and that this truth can be discovered through careful,
objective, impartial use of the experimental (empirical) method. By contrast, post-
positivist approaches, influenced by post-structuralism and post-modernism, contend
that, in the modem world, values, opinions, and truth itself are constantly shifting
and changing to accommodate the shifts in power structures and changes in culture
and language. Post-positivist approaches contend that there is no such thing as an
objective, single Truth and that social science should concentrate on exploring the
plurality of ways in which individuals and groups construct their understanding of
the world depending on cultural and local context. Post-positivist theorists often
advocate the use of qualitative methodologies methodologies that are concerned
with individuals and groups understanding of their context and are typically
focused on language and narrative such as ethnography, discourse analysis, and
narrative analysis. This paper will employ the narrative approach to explore what
(and how) is being said about the siting of LNG terminals.

Emery Roe and Narrative Policy Analysis
Policy analysis has traditionally employed conventional quantitative
approaches utilized in social sciences in an attempt to provide policymakers with
concrete answers to questions of which policy will deliver the most benefit for the
least money. This places policy analysts in an interesting conflict while, in the
world of policymaking, the analyst is expected to provide practical, objective
solutions, in the world of academia, a policy scientist is expected to be able to
conform, to an extent, to paradigms that dominate academic discourse. In the past
several decades, these conflicts have centered on the differences between post-
positivist and positivist approaches; in the field of social sciences the former have
become accepted as a legitimate way of doing science. Most accept that language
and culture influence political decisions and that it is no longer possible to describe
policy choices as always objective and rational. This has led to a growing interest in
applying qualitative methodologies to the field of policy analysis. One of the most
prominent theorists to attempt to introduce qualitative methods, specifically narrative
analysis, to policy sciences, is Emery Roe, who argues that
Stories commonly used in describing and analyzing policy issues are a force
in themselves, and must be considered explicitly in assessing policy
options. Further, these stories (calledpolicy narratives...) often resist
change or modification even in the presence of contradicting empirical data,

because they continue to underwrite and stabilize the assumptions for
decision making in the face of high uncertainty, complexity, and
In this conception, uncertainty is seen as lack of knowledge about the
policy issue, complexity is defined as the issues intricacy, and polarization is seen
as the concentration of groups around extremes in the issue. Roe contends that
there are public policies that are so uncertain, complex, and polarized that the only
thing an analyst can do is to explore the stories that the stakeholders tell in order to
understand the causes of this uncertainty and complexity.63 One could argue that
LNG siting is such an issue, as evidenced by the fact that the public fears about LNG
persist and multiply even in the face of empirical data that seem to underscore the
safety of the technology and the economic data that suggests the desirability of
natural gas importation to the U.S. economy. This might be precisely the kind of
policy issue where asking, Whats the story? Who is telling it? might be a useful
policy analysis tool.
Roe sees narrative as a story that can have a scenario structure (a traditional
story structure that has a beginning, a middle, and an end) or a structure of an
argument, with premises and conclusions. Of interest here are policy narratives
that dominate the issue in question and, as a result, are seen by the stakeholders as
62 Roe, Emery. Narrative Policy Analysis: Theory and Practice. (Durham: Duke
University, 1994), 2.
Ibid., 3.

underwriting (that is, establishing or certifying) and stabilizing (that is, fixing or
making steady) the assumptions for policy making in the face of the issues
uncertainty, complexity, or polarization.64
In addition to such dominant stories, Roe also describes nonstories,
counterstories, and metanarratives. Nonstories are narratives that lack traditional
story structure (such as circular arguments or arguments without beginning, middle,
and end), while counterstories are narratives that are set in opposition to the
narratives that dominate the issue. Metanarrative is a story, sometimes generated by
the analyst herself, that reframes the issues in such a way as to enable it to emerge as
a policy problem that can be analyzed using traditional tools of policy analysis
(economics, law, statistics, etc.) by underwriting and stabilizing the assumptions for
decision making on an issue whose current policy narratives are so conflicting as to
paralyze decision making. For Emery Roe, the goal of narrative policy analysis is
generating such metanarratives in situations where the goals and desires of various
stakeholders groups are so oppositional that compromise is impossible. A
metanarrative makes it possible for an analyst to generate a new story on the issue so
as to be able to find a policy solution.65
64 Ibid., 3
65 Ibid., 3-4,55-

By new story the author means not a story that forces consensus on the
parties and obscures conflict, but a story that presents a different agenda that
attempts to include the main opposing arguments/scenarios of the stakeholders,
without dismissing any of these points, however extreme they may appear. A
metanarrative, of course, is not always possible. A countemarrative can also displace
the dominant narrative. Dominant narratives often maintain control over debate
about a policy issue, despite being refuted or discredited by scientific findings,
because discrediting the argument for action does not remove the policymakers
perceived need to act. In that case, in order to displace the refuted narrative, the
countemarrative needs to tell a better story, one that is equally straightforward.66
At this point, it might be useful to provide an illustration of how Roes
methodology can be applied to a real policy problem. The example of the California
Medfly Controversy of 1980-1982 might be particularly useful, as it deals with a
hazardous and undesirable substance being introduced into a community. In this
case, California Department of Food and Agriculture (CDFA) had to quickly design
and implement a program to deal with an outbreak of medflies in fruit growing
communities in the Los Angeles and Santa Clara counties. While the program of
stripping trees of fruit, releasing sterile medflies, and applying pesticides was
successful in Los Angeles, the infestation continued to spread in Santa Clara,
covering over 200 square miles. This led to calls for abandoning the ground-based
66 Roe, 40, 52.

program and moving to a mass-scale aerial spraying of the area, the proposal
opposed by several environmental groups and by local residents. The aerial program
was eventually implemented and applied to 1,300 square miles.*7
In this case, uncertainty resulted from the lack of clear-cut scientific evidence
in favor of either ground-based or aerial-based program, while the controversy arose
because the stakeholders fell into two camps: those supporting the aerial program
(Aerial Story) and those opposing it (Aerial Critique). In Roes analysis, Aerial
Story emerged as one with a conventional story format, that is having a beginning,
middle and end, which explained the threat of medflies to the agriculture industry,
provided evidence of success for similar aerial programs, explained why expanding
the ground program was not feasible financially, and provided an implementation
plan for aerial spraying with the final goal of eradicating the medfly. On the other
hand, Aerial Critique provided only a critique of aerial spraying, but no real stoiy of
its own: those opposed to aerial spraying were much more vocal in expressing their
opposition to aerial spraying than they were in describing their preferred policy
choice of expanding the ground-based program. This was largely due to the fact that
the vocal opposition to aerial spraying consisted largely of environmental groups,
who found it problematic to argue in favor of expanding the ground-based program
because it used pesticides that were more hazardous than those used in aerial
spraying. Furthermore, Aerial Critique found itself sidetracked into arguments over
Roe, 55-56

various smaller controversies surrounding the issue (Should the medfly be eradicated
or controlled? Were the officials too slow to respond? Were new ways of detecting
medfly infestations needed? Was the release of sterile medflies methodologically
sound enough to be used as an alternative to ground spraying?), which provided
conflicting factual information, increased uncertainty, and made the Aerial Critique
appear incoherent and hard to read as a story. It has to be noted that that doesnt
mean that the Aerial Critique was without merits or altogether unsupported by
scientific evidence; however, these features made it less readable as the basis for
action and vulnerable to the Aerial Story proponents who disparaged it as a set of
disparate complaints from skeptics having little in common but their opposition to
aerial spraying.68
In the end, Aerial Story came to dominate the policy debate, as it had the
support of politicians and agribusiness interests in the area. In addition, after the
initial opposition from the media, the program has garnered support and positive
coverage after its implementation and attracted more positive testimony from
politicians and CDFA and USDA officials. In the face of uncertainty about factual,
68 Ibid., 56-64.

scientific merits of either policy choice, Aerial Story was able to become the policy
choice because it had a coherent structure and was told by key stakeholders in the
According to Roe, policy narratives are a way for stakeholders to transform
issue uncertainties and complexities into perceptions of risk. Risk, in this
conception, is the product of the severity of the hazard associated with an event and
the probability of that event occurring. In the case of the Medfly Controversy,
hazard is negative effect on health from the aerial spraying of pesticides.
Supporters of the Aerial Story used government reports to form an opinion that the
risk of this particular hazard was low. Proponents of the Aerial Critique never moved
past other issues raised by the Aerial Story to address the validity of the figures used
to estimate the health risk from aerial spraying. If this issue was raised, however, a
metanarrative could have been constructed that would explore the possibility of
redesigning the aerial program to reduce its level of perceived risk to the level of
risk thought associated with the ground-based program. By not bringing to the table
an alternative policy choice and by not concentrating on the issue that could bring
about a metanarrative, Aerial Critique increased uncertainty. According to Roe, this
is often the case in situations such as the Medfly Controversy, in which asymmetrical
narratives that is, situations where one side tells a story while the other provides
only a critique of that story, but no narrative of its own are present. Asymmetrical 69
69 Ibid., 56,65.

narratives often arise in the presence of economic and political power differences (in
this case, the Aerial Story had the support of the USDA, agribusiness interests, and
prominent politicians, while Aerial Critique was told by disparate local groups).70
Asymmetrical narratives will be discussed in more depth elsewhere in this
work, as they become pertinent to the analysis. At this point, however, it is necessary
to note that Roes methodological suggestions do not address all of the aspects of
this project. As this project will explore the content of the stories stakeholders in the
problem of LNG terminal siting tell on their Internet websites, a framework is
needed that will account for the unique features of stories told on the Internet. Thus,
websites use both text and symbolic artifacts, such as images and graphics, to tell
their stories; such symbolic artifacts are not accounted for in Roes framework, but
might have a powerful story to tell. In addition, Roes suggests that his framework
can be applied in an informal way that involves identifying stories, nonstories and
metanarratives subjectively, from transcripts of interviews with the stakeholders (the
analyst would simply ask herself if the story has a beginning, middle, and end;
identify nonstories that do not have those narrative features; and construct a
metanarrative) or in a highly structured process that disaggregates the interview
transcripts into discrete problem statements that assert a causal relationship
between events discussed. These statements from all interviews are then used in a
70 Ibid., 68-72.

complex data modeling procedure that creates frequency tables that can then be used
to identify problems that are mentioned most often, as well as to create a network of
purported causal relationships said to exist among problems identified by
interviewees. This network analysis can then be used to reaggregate multiple
interviewee statements in order to identify and describe the story that the participants
are telling.71 While the first approach is highly subjective and doesnt satisfy this
authors desire for a more rigorous methodology, the second approach involves
multiple interviews with the stakeholders, during which the researcher has an ability
to ask probing questions that explore the cause-and-effect relationships present in the
situation. This project, however, utilizes the stories that have already been told and
cannot further probe the narrators. Therefore, in order to be able to explore symbolic
data, such as images, and to take a fairly structured methodological approach that
would be more suitable to the nature of data to be used, this work will utilize Roes
framework together with the framework that Dvora Yanow calls an interpretative
approach to policy analysis and the rhetorical criticism approaches discussed by
Roderick Hart and Suzanne Daughton.
Dvora Yanow and Interpretative Policy Analysis
Much like Roe, Dvora Yanow starts with the premise that policy analysts
need to understand the influence of values and beliefs on policy decisions, as well as
the ways in which target populations interpret policies based on their context. In this
71 Roe, 155-161.

framework, neutral, objective science is not possible, because policies have
multiple meanings (in contrast to facts emphasized by traditional policy analysis).
All living, writes Yanow, requires sensemaking, and sensemaking entails
interpretation; therefore, policy analysis also requires interpretation. Yanows
theoretical framework is based on hermeneutics, which contends that individuals,
with the help of prior knowledge, project meaning on cultural artifacts (including
literature, language, art, actions) and that these artifacts can be then studied in order
to extract those meanings. Therefore, what she calls interpretive policy analysis is an
approach that explores the meanings that policies have for a broad range of policy-
relevant publics or the what and the how of those meanings. And, much like
Roe, Yanow wants to concentrate on the plurality of possible meanings and the
tensions between them in order to discover how policy issues are being framed by
the affected parties. That is, interpretive policy analysis focuses not only on the
meaning of policies and the values, feelings, and beliefs they express, but also on
which issues are being highlighted and which are being obscured and how, through
the use of language, action, and symbols.72 Language and symbols can be analyzed
using rhetorical approaches.
72 Yanow, Dvora. Conducting Interpretive Policy Analysis. Qualitative Research
Methods Series 47. (London; Sage, 2000), 3-8,11-14.

Rhetorical Criticism
Roderick Hart and Suzanne Daughton define rhetoric as an art that creates a
story...[using] common ideas, conventional language, and specific information to
influence audiences feelings and behaviors. This story always has a purpose. Given
this definition, a rhetor has five tasks: to force change by using symbols rather than
violent acts; to present her/himself as a helper rather than an exploiter; to convince
the audience to make new choices', to narrow the available options to a few choices;
and to be subtle by not specifying the details of the policies advocated. The job of
the rhetor, therefore, is to sell choices to an audience that is resistant to change and to
making choices.73 One could argue that, in this definition, at least one feature of the
rhetors job is to deceive the audience; that is certainly a popular belief when it
comes to political rhetoric. However, it might also be argued that there could be
other reasons for not providing extensive detail, such as assumptions about
knowledge level of the audience, whether the audience shares the convictions of the
rhetor (e.g. a campaign speech delivered in front of supporters), and simply the need
to not bore ones audience.
A message is rhetorical if it has the following three features: delineation of
the good...resonance for a particular audience...and clear or clearly implied policy
recommendations. A rhetorical message has several features: it unburdens, by
73 Roderick Hart and Suzanne Daughton, Modern Rhetorical Criticism, 3rd ed. (New
York: Pearson, 2005), 7.

allowing the rhetor to champion her/his issue in front of an audience; it distracts, by
sidetracking the audience attention to focus it entirely on the rhetor; it enlarges, by
reframing issues and presenting new solutions; it names, by creating a new
vocabulary to talk about ideas; and it empowers by giving ideas a voice and a vehicle
for transmission. We can evaluate whether a piece of rhetoric was successful at
fulfilling this function using several standards, a few of which are relevant to this
work: the Utilitarian Standard (Does the message do what the rhetor intended it to
do?), the Scientific Standard (Is the message based in fact and are the conclusions
warranted given available evidence?), Political Standard (Does the message advance
the social causes and goals endorsed by the critic? Will any groups be harmed by the
message?) and the Psychological Standard (Does the message unburden the
emotions of the rhetor? Which fears does it address? What is the effect on the
audience?). When we evaluate rhetoric according to these standards, we analyze
ideas to: 1) explore what is present and what is omitted; 2) uncover rhetorical
patterns (language); and 3) categorize rhetorical tone (emotion).74 Three specific
evaluation methods will be discussed in the chapter on methodology.
74 Hart and Daughton, 12,14-18, 34,60-61.

Ideas in the Public Sphere
...The responsibility of policy analysts, writes Robert Reich, is not only to
choose the best means of achieving a given objective. It is also to offer alternative
ways of understanding public problems and possible solutions, and thus to expose
underlying norms to critical examination. An analyst might choose to critique
widely held, yet problematic beliefs about policy issues; to reframe our
understanding of events that drive policy; to look at similarities in ways in which the
public makes meaning of different policies; to present new ideas for the future; and
to question the conventional metaphors and analogies used to justify and explain
policies. To accomplish these goals, an analyst must examine what Reich calls
public ideas, or ideas discussed in the public sphere and in the spirit of democratic
debate. Such ideas are increasingly important in the age of media, pluralism, and
expert opinion.75
Despite the widely spread belief that the lay public is interested primarily in
maximizing personal utility and is content to leave the creation of public policy to
the invisible hand, theorists such as Steven Kelman believe that the public is
interested and involved in the creation of public policy.76 Others, such as Mark
75 Reich, Robert. Introduction to The Power of Public Ideas. (Cambridge: Harvard
University Press. 1988), 6,10.
76 Steven Kelman, Why Public Ideas Matter, in The Power of Public Ideas, ed.
Robert Reich (Cambridge: Harvard University Press, 1988), 31.

Moore, contend that public ideas guide public establishing the context
within which public policy is debated and executed. In doing so, public ideas that
come to dominate a policy area not only inform the beliefs and values that drive
policy, but also prescribe the appropriate form of public action and even authorize
and instruct different communities to take public action on a policy issue.77 Thus,
public ideas determine our understanding of public good, inform our
understanding of appropriate policy solutions to social problems, and drive our
involvement in political and public action.
To become public, ideas must, however, reach the community. Up until
very recently, print and television media served as a vehicle for disseminating public
ideas. Martin Linski notes the centrality of the media to the communication between
public officials and the public when he lists the variety of ways in which policy
makers communicate with the public: press releases, press conferences, speeches
covered by the media, and leaked information are all important tools in making
policy ideas public. At the same time that the importance of the press in public
debate grows, the press itself is experiencing an increase in commercialization and a
consolidation. The number of cities with more than one newspaper has been
dwindling, thus limiting the publics access to alternative ideas. In addition, mass
media is not simply an objective observer of the public process, but an active
77 Mark Moore, What Sort of Ideas Become Public Ideas, in The Power of Public
Ideas, ed. Robert Reich (Cambridge: Harvard University Press, 1988), 75.

participant that not only reports on policy issues, but affects and changes the context
and content of public debate by expressing a standpoint.78
While mainstream mass media is seen as increasingly commercialized and
consolidated (both in terms of ownership and in terms of adhering either to a
liberal or conservative standpoint), the Internet is becoming a more attractive
vehicle for expressing political ideas, as seen in the recent explosion in the
popularity of political blogging. One might wonder, however, whether the Internet is
mainstream enough to be used by policy stakeholders to disseminate their stories
and to be seen as a legitimate source of narrative data by analysts. In fact, it appears
that not only has the use of the Internet grown over the past several years, but that it
is increasingly being used by Americans as a source of political information.
According to the Pew Internet and American Life Project, 63 percent of adult
Americans, or 126 million, use the Internet. Between 2000 and 20002, online news
consumption grew by 50 percent (from 52 million Americans to 78 million) and the
number of Americans searching for political information and news online grew by
57 percent, from 30 million to 47 million.79 Yet another Pew study found that the
role of the Internet in politics had its breakout year during the last presidential
78 Martin Linski, The Media and Public Deliberation, in The Power of Public
Ideas, ed. Robert Reich (Cambridge: Harvard University Press, 1988), 205-208, 209-
79 Madden, Mary. Pew Internet and American Life Project. 2003. The Changing
Picture of Whos Online and What They Do. Washington, DC. (accessed March 27, 2006), 1-8.

campaign, when 75 million Americans (37 percent of American adults and 61
percent of U.S. online population) used the Internet to find political news and
information, talk about candidates and issues in emails, and contribute to campaigns
The study raised two interesting points: the majority of Americans who get
their political news online said that the information they found on the Internet helped
them decide how to vote (52 percent) or encouraged them to vote (23 percent) and a
significant number said that they used the Internet for political purposes because they
couldnt get all the information they needed from traditional media sources or
because the Internet offered information that is not available anywhere else (44%).80
These data indicate that the Internet is becoming an increasingly important source of
political information and that information obtained on the Internet can sway public
opinion, encourage some to engage in political action, and find an audience for
viewpoints that are not always given coverage in mainstream media.
The importance and mainstreaming of the Internet in the political arena can
be observed in the growth of e-Govemment. Agencies are increasingly sharing
information about policies with the general public and inviting public participation
through e-filing and email. A community resident worried about proposed
construction of a LNG terminal in her/his area can easily obtain safety reports,
80 Rainie, Lee, Cornfield, Michael, Horrigan, John. Pew Internet and American Life
Project. 2005. The Internet and Campaign 2004. Washington, DC. (accessed March 27, 2006), 1-3, 24.

environmental impact statements, legal information, Congressional testimony, and
information about specific actions a member of the public can take (complete with e-
filing of some forms) on the FERCs website dedicated to LNG. In this way,
agencies are not only providing public with access to information, but are also
attempting to influence public opinion on policy issues.
For the purposes of this study, one of the main advantages of using Internet
narratives, in the absence of personal access to stakeholders, is that they are entirely
produced by the stakeholders and are not mediated by outside standpoints (as
opposed to stories in mainstream media that are authored by journalists and mediated
by their beliefs, values, and stated opinions). The Internet allows the stakeholders to
express their ideas directly to the public and to expand on their argument more fully
than they could within the confines of traditional media.

Dvora Yanow suggests that there are at least three communities of meaning
in every policy situation: policymakers, implementing agency personnel, and
affected citizens or clients. Communities of meaning are formed where individuals,
through their belonging to a group, come to share interests, political views,
professional membership, organizational structure, etc, so that the groups
cognitive, linguistic, and cultural practices reinforce each other, to the point at
which shared sense is more common than not. Communities of meaning are fluid
and complex there is more than one in every organization.81 For our purposes,
communities of meaning are defined as: regulatory agencies, LNG industry, and
opposition groups (i.e. citizen groups that oppose siting of LNG terminals in their
Several websites were chosen and classified into each of these communities.
Federal Energy Regulatory Commissions (FERC) LNG website was chosen to
represent the regulatory agency community, as it is the lead agency on LNG siting
and has a comprehensive website devoted to LNG. Three websites promoting LNG
projects were chosen to represent LNG industry, one that has obtained FERCs
81 Yanow, 10

approval (Weavers Cove, proposed by Hess LNG, in Fall River, MA), one that is
awaiting approval (Broadwater, proposed by TransCanada Corporation and Shell in
Long Island Sound, NY), and, for comparative purposes, a website for an existing
terminal (Cove Point, owned by Dominion Energy, currently proposing an
expansion). Four websites were chosen to represent opposition to these proposals:
two created by the opposition to the Weavers Cove project (one community group
website and one environmental organization website) and two created by opposition
to the Broadwater proposal (again, one community group website and one
environmental organization website). This selection was made in order to compare
the community group narrative to the environmental group narrative in each case and
to provide some cross-comparisons (i.e., community narrative in Fall River to
community narrative in Long Island and environmental narrative in Fall River to
environmental narrative in Long Island). In this conception, community group simply
refers to a group of concerned citizens who have come together around the single
issue the proposed siting of the LNG facility in their community and with a
single goal, to oppose this facility (by contrast, environmental groups in these cases
have existed before the siting controversy arose and the siting itself is typically one
issue among many on their agenda). The fifth website is created by a New England
environmental group, Conservation Law Foundation which, while not opposed to
LNG importation and technology, is opposed to the terminal siting policy as it is

currently designed and implemented. This narrative was chosen because it appears to
provide an emerging metanarrative on the issue of siting LNG terminals.
Three methods for analyzing rhetoric, as outlined by Hart and Daughton,
were used: Hart and Daughtons judgmental approach and analysis of rhetorical
evidence or clarification devices, and Stephen Toulmins system for analyzing
informal arguments which is similar to Roes disaggregation technique. Each of
these methodologies requires the narrative to be disaggregated into statements
(usually containing one or two sentences) that are then codified into categories
outlined in each method.
In Hart and Daughtons judgmental approach, rhetoric is analyzed from the
perspective of the audience by concentrating on the judgments listeners are asked to
make during persuasion. The approach is based on the preposition that rhetors
assume that the audience will make judgments about their arguments and that the
text, therefore, is a record of the kinds of judgments audiences are being asked to
make. This approach lists four pure judgmental requests. Factual requests are
statements that are treated as common knowledge/fact that ask listeners to consult
the world around them and consider what is true. Factual requests are not judged by
scientific standards, as they merely invite the listener to consult her/his
understanding of the world when judging the rhetors statement. Desirable requests
ask listeners to consult their own general wishes and preferences (or those of their
social groups) to judge whether the proposition would make them happy or

unhappy in matters of taste, efficiency, beauty, or practicality. Adjudicative
requests ask listeners to consult some formal code of behavior and consider how a
particular behavior measures up to such codes, which are more strident than the
Desirable standards and can include law, religion, or ideology. Finally, directive
requests explicitly or implicitly detail the positive consequences of adopting a new
attitude or standard of behavior and ask the audience to approve the means the
rhetor intends to utilize in obtaining those consequences. In addition to these pure
judgments, rhetorical texts also contain mixed messages (factual/desirable,
factual/adjudicative, factual/directive and desirable/directive). The task of the analyst
is to catalog the statements found in the text according to those types of judgments,
while asking the following five questions: 1) Which judgmental requests are present
in the text? 2) Where are they located in the text? 3) Which judgmental requests are
absent or underrepresented? 4) What are the implications of how the judgmental
requests are clustered and positioned? 5) How is the audience expected to respond to
these requests? This approach is especially useful in exposing rhetorical patterns and
in helping the critic highlight the ideological characteristics of the message, as
desirable judgments are usually based on informal beliefs, whereas adjudicative
judgments call on institutionalized (ideological) beliefs. 82
82 Hart and Daughton, 67-69,71-72.

Rhetorical evidence and clarification devices are techniques that make the
argument more persuasive. Hart and Daughton suggest that, by analyzing the
presence of such devices in the text, the analyst can evaluate the effectiveness of the
argument by exploring: 1) whether the argument is flatly assertive vs. reliant on
evidence; 2) whether the argument is sketchy and broad vs. specific, with less
propositions but more evidence; 3) which arguments are supported by evidence
and which arent and why; and 4) what kinds of evidence are used and how are these
evidence types positioned in the text. Serial examples add totality to a speakers
remarks by presenting, in scattered fashion, numerous instances of the same
phenomenon. Extended examples add vivacity by painting a detailed picture of a
relevant event. Quantifications add substance to the text by presenting concrete
evidence. Isolated Comparisons add realism by drawing an analogy with the past
experiences of the audience. Extended comparisons add psychological reference
points by structuring...perceptions along familiar lines. Testimony adds
inclusiveness to the argument by quoting appreciatively from respected sources
and depreciatively from derided sources. Definition adds to the specificity of the
text by presenting opposed elements. Finally, contrast adds dramatic quality by
depicting opposed elements.83
83 Hart and Daughton, 83-85

Finally, Stephen Toulmin proposes a reduction of text to an outline (or
skeletal structure) in order to establish...overall logical movement.84 The system
allows the analyst to depict arguments as graphs, similar to Roes disaggregation
technique, or to simply create a record of the main features of the text: Major Claims
(MC), Major Data (MD), and Warrants (W). Major Claims are the broadest, most
encompassing statements that are frequently repeated or restated in the text, are at
a higher level of abstraction than other statements, and are meant to be remembered
when the details of the message have been forgotten. Major Data lie at the level of
abstraction immediately beneath that of the Major Claim and are the supporting
structures of the discourse or facts and devices such as images and graphs that
support the argument. Warrants make the movement from Major Data to Major
Claim possible. They are hypothetical statements which...authorize the sort of step
to which our particular argument commits us. For instance, in the example given by
the authors: This politician is a Christian [MD] Christians make good
officials [W] -> Lets vote for him [MC]. Warrants can be substantive (based on
what is considered to be factual knowledge), motivational (claiming that some
desirable end must be achieved being endangered) or authoritative (based on
perceived credibility of the speaker or the source of the speakers testimony).
In examining the text using Toulmins method, the analysts poses the
following five questions: 1) What are the Major Claims, particularly those that are
84 Ibid., 93.

being restated? 2) What are the Major Data? 3) What are the range of warrants that
could reasonably authorize such Data Claim movements? 4) Are these potential
warrants substantive, motivational, or authoritative? 5) Which or the warrants were
present/absent in the text? The last question is of particular importance, as some
warrants remain unstated, thus inviting the audience itself to authorize the movement
from Major Data to major Claims. Looking for the missing and the unstated in the
text allows the analyst to achieve an understanding of the subtle ways in which
discourse invites the audience to make meaning of the narrative.
This approach allows the critic to surmise whether the rhetor expects the
audience to share her/his ideology (we can suggest that to be the case if warrants are
not provided or are suppressed); to explain the tone of the message (for instance, a
text that moves rapidly from major data to major data in the service of one major
claim would have a bold tone); to explain why the message fails to persuade the
audience (such as a message with many unconnected major claims or a message with
major claims not directly followed by major data, both of which can come across as
rambling); and to evaluate the text while deemphasizing its beguiling use of
85 Hart and Daughton, 92-95,97-98

This section will present an analysis of narratives using the methodologies
outlined above. It is expected that analysis of clarification devices will allow the
author to critique the persuasive quality of the narratives (i.e. whether or not they are
successful in convincing the audience of the truthfulness of rhetors statements about
LNG terminal sitings), while the judgmental and the Toulmin approaches are
expected to help uncover content and themes, and logical flow and tone of the
narrative, respectively. In doing so, the following sections will explore the structures
of the narratives, catalogue their major themes, create the summary of the content of
each narrative using those themes, and discuss, where appropriate, the narrative tone,
use of language, and use of evidence in support of major themes. The goal of this
analysis is to uncover the content and the structure of the narratives and thus to
identify whether the narrative in question is a story (i.e. has a beginning, middle, and
end), a counterstory, or a nonstory (e.g. a critique). The methodology, when applied
to data, is expected to help answer questions such as: Does the content of this
website tell a story, a counterstory, or a nonstory? How can we tell? Can any of these
stories be seen as dominant, i.e. do they drive the policy choices involved in the

siting of LNG facilities? Can the counterstories supplant the dominant stories in this
debate? Is there an emerging metanarrative?
FERC Story: Informative-Authoritative Narrative.
This section presents an analysis of the narrative created by FERC,86 which is
of interest because FERC is the lead agency on LNG terminal siting and on setting
siting policy and, as such, can be expected to provide a narrative driven by factual
data in the service of the goal of persuading the audience that FERCs siting policies
are appropriate.
Indeed, the FERC website, Liquefied Natural Gas (LNG) in the U.S. is a
highly detailed information site that appears to have a lay audience in mind. It has
four sections Safety and Inspections, Environment, Industry Activities, and
General Information each with several subsections and containing simple
presentations of information about LNG technology, markets, and uses. The website
design is uncluttered, easy to navigate, and efficient in its presentation of factoids,
FAQs, maps, calendars, contact information, and even a detailed outline of the ways
in which members of the public can get involved during the various stages of
FERCs review of proposed LNG terminals again leaving one with an impression
of an information site designed for the use by lay public. The website also contains a
86 Federal Energy Regulatory Commission. Liquefied Natural Gas (LNG) in the
U.S., Federal Energy Regulatory Commission, (accessed, March 1,2006).

photo gallery which displays photographs of LNG vessels at sea, docking, and
entering ports. LNG facilities photographed dont have a typically industrial look,
but are pictured as new and clean, invoking feelings of efficiency. Most of the
photographs feature U.S. Coast Guard vessels with clearly visible USCG signs near
the LNG vessels and there are several photographs of USCG inspectors at work,
invoking feelings of safety and professionalism.
Overall, safety is a major feature of the website, with emphasis on FERCs
dedication to safety repeatedly restated, and safety reports and environmental impact
statements readily available and linked from several sections of the website. One
could say that the overall impression one takes away from visiting the site is that of
FERCs professionalism, transparency, dedication to public safety and public
involvement, and authority in the field of LNG and terminal siting.
In terms of narrative structure, FERC is telling a story about LNG (what it is,
why it is needed, and how it gets to the consumers) and FERCs regulatory
involvement (laws that empower FERC to act as a lead agency and what it is that
FERC does in this role), although due to the nature of website design, the story is not
presented in a linear manner, as one moves from one section to another.
The overall narrative is, to an extent, intertextual, a narrative feature that Roe
defines as a condition where there is no primary Text the importance of which
supersedes all other narratives and which provides the main definition of the
problem. However, unlike Roes observation of high intertextuality of national

budgets, which depend on previous budgets to define the terms and items in the
current budget,87 FERC, despite linking to other agencies and studies and repeatedly
redirecting users to other websites (such as the U.S. Department of Energy, Energy
Information Administration, Office of Pipeline Safety, and legal sites) for more
information, is clearly establishing itself as the primary narrator (not surprising, since
FERC is the lead agency in LNG regulation). This tipping of the balance of primacy-
intertextuality in favor of primacy of FERCs narrative, combined with yet another
feature noted by Roe the absence of named authors establishes FERC as the
Grand Author, a narrative feature which Roe calls command model, in which one
entity (treasury, the president, or parliament, etc.) determines the primary or
dominant reading.88 As a result, one could call this particular story an Informative-
Authoritative Narrative, as FERC simultaneously informs the public about main
issues in LNG production, usage, and regulation and establishes itself as a primary
authority in the field, thus positioning its narrative as the dominant story about LNG.
FERCs credibility and authority are also established through the use of
several rhetorical techniques. Analysis of judgmental clusters in the text shows that
Factual cluster is by far the most developed of the eight possible clusters, closely
followed by Adjudicative and Factual!Adjudicative clusters. Factual statements (i.e.
statements that are treated as commonsense facts but do not necessarily meet
87 Roe, 25-27
88 Ibid., 32

scientific standards) occur twenty-five times throughout the text and address two
areas: 1) FERCs mission and activities (e.g. Public safety is among FERCs
highest priorities when it comes to regulating LNG terminals, FERC uses a
comprehensive siting process, FERC ensures the safety and security of LNG
terminal projects from the time they are conceived...and during their construction
and operation, Since the 9/11 attacks, FERC has formed a new LNG Engineering
branch...devoted exclusively to safety and security, FERC staff visited Skikda,
Algeria to learn more about the accident at this LNG export terminal); 2) LNG-
related information (e.g. LNG is economically viable at todays market price,
About 96 percent of the worlds proven natural gas reserves are outside of North
America, In its liquid state, LNG is not explosive, and Only four accidents have
occurred over a 60+ years). These clusters are positioned in areas of the text
dedicated to LNG uses and safety and FERCs regulatory activities, thus inviting the
reader to increase her/his knowledge of the technology. One could argue that the
Factual clusters are designed to minimize fears (e.g. those arising from the belief that
LNG is explosive) by presenting information about LNG and the industry safety
record, and to increase the trust in FERC as the main decision-maker on the siting
issue by describing FERCs regulatory activities in a transparent, often step-by-step,
manner. It has to be noted, however, that FERCs statements sometimes do not meet
the scientific standard while FERC contends that only four accidents have

occurred in the past 60 years, other information presented in this work indicates that
more accidents have, in fact, occurred.
As mentioned above, the Adjudicative and the Adjudicative/Factual clusters
are fairly well-developed. Adjudicative statements invite the reader to evaluate
information in light of institutional standards, while adjudicative/factual statements
rely on the credibility of another institution to present the policy as acceptable. Here,
adjudicative statements invite the audience to evaluate FERCs regulatory activities
(such as the authorizing the siting of the facilities and imposing additional safety
and security measures on a proposed LNG terminal to ensure and enhance safety and
system reliability) in light of acts that authorize FERC to act as a lead agency, while
the adjudicative/factual statements borrow the credibility of statutes that ensure
environmental protection, federalism, and international marine safety, such as Clean
Water Act, Coastal Zone Management Act, Clean Air Act, etc., to highlight the
acceptability of FERCs siting policies in light of standards set by legal codes. Both
clusters establish FERCs authority (granted by acts and laws) and accountability
(answerable to these laws), as well as interconnectedness of FERC with other
agencies and entities (states, USCG, etc.), all working together in the service of
public interest.
Interestingly, the Desirable!Directive and Factual!Directive clusters, which
are used as persuasive approaches and as techniques that invite the audience to
transfer the benefits of a policy to their own lives, are not present. This indicates that

the intent of the FERC site is not to create a sales pitch, but to present the evidence
that will lead the audience to conclude that LNG is necessary to the U.S. economy
and consumers. For instance, the Directive cluster is not very developed and occurs
only in the Importance of LNG section. While not very developed, it utilizes a
bullish language (growth in U.S. natural gas supplies will depend on non-
conventional domestic production, natural gas from Alaska, and imports of LNG
and in order for the US to meet its increasing demand for natural gas, LNG must
become an increasingly important part of the US energy mix, emphasis added), the
tone of which indicates that LNG imports are a done deal and invites the audience
to agree with FERC that LNG is a necessary choice. The fact that the Desirable
cluster is not well-developed adds weight to this conclusion, as the audience is not
asked to consult its own preferences and wishes on the use of LNG and natural gas.
The message sent by the site can be summarized as LNG is here to stay and grow,
which makes a hard-sell rhetorical approach unnecessary.
Analysis of Clarification Devices reveals that FERCs narrative, despite not
utilizing a hard-sell approach, is very persuasive nonetheless, as it makes extensive
use of supporting materials and clarification devices, particularly quantification and
testimony, to make a few arguments supported by extensive evidence. While FERC
defines LNG in detail, definitions are scattered throughout the text, instead of being
clustered in the same section, thus adding specificity to the text without
overwhelming the audience with the technical information (again, creating the

impression that FERC has a lay audience in mind). LNG is defined as a liquid (by
contrast with gaseous state), as a natural gas (by contrast with gasoline), and as a
substance that is not explosive (by contrast with conditions that can produce an
explosion). Many other concepts, however, are not defined (e.g. safety, security,
states rights), thus indicating that the audience is expected to share cultural
assumptions that help us define such concepts. Mentioned but not defined are
controversial issues such as question over who has primary jurisdiction over LNG
in California, the State or FERC, Mexican nationalization issues, and NIMBY
concerns, with the lack of definitions (or other clarification devices in the Public
Concerns section) taking on an appearance of an obfuscation or at least a telling
silence on these issues.
The text makes extensive use of quantification, with dates, numbers, costs,
and percentages appearing fifty times throughout the text. Evidence is presented to
add substance to the arguments about U.S. need for natural gas and LNG imports
(the US is consuming about 25 percent of the worlds annual natural gas
production and EIA estimates that LNG could account for as much as 21 percent
of the total U.S. natural gas supply in 2025); safety record of the industry and safety
of the technology (e.g. in the past 40 years there have been more than 33,000 LNG
ship voyages without a significant accident, Only four accidents have occurred
over a 60+ years, and natural gas is only flammable within a narrow range of
concentration in the air (5% to 15%)); and evidence that supports the claims that

FERCs regulatory activities are extensive and proactive (there are about 40 LNG
terminals that are before FERC, about 80% of currently proposed LNG projects go
to prefiling, there are 16 facilities under FERC jurisdiction in the continental U.S,
the Commission has articulated its new LNG policy in the December 18,2002
Hackberry policy, as well as 20 other dates of important LNG-related activities that
involve FERC). Thus, while the amount of supporting quantifiable evidence is
extensive, it can be clustered along three claims LNG is needed, the safety
record of the industry is satisfactory, and FERC is proactive in ensuring proper
regulation and public safety. However, there is a telling silence, again, on the issue
of public concerns and issues mentioned above that FERC describes as
controversial, such as the dangers of LNG vapor clouds or gas quality issues,
which are not substantiated by numbers and dates.
The text makes some use of serial and extended examples, with the former
utilized five times to briefly present complex issues (such as security issues, e.g.
examples of these requirements include security patrols, protective enclosures,
lighting, monitoring equipment, and alternative power sources) and the latter used
nine times to expand on safety and security measures, LNG accidents, and FERCs
review/approval/post-approval involvement in LNG terminal siting and operations.
Both devices are used to present information and reiterate FERCs proactive
regulatory activities and commitment to public safety by presenting a detailed picture
of FERCs activities. Similarly, the text makes some use of testimony, usually when

presenting economic data (e.g. World natural gas reserves are abundant, but located
far from consuming markets, followed by a link redirecting the reader to the Energy
Information Administration site) or safety data (e.g. links to the Sandia Report), that
is, when dealing with issues on which other agencies and organizations might have
more expertise than FERC. Thus, testimony is used to support two of FERCs
recurrent arguments: LNG is needed by the U.S. economy and LNG is safe.
However, it has to be restated that this is not a hard-sell, and these arguments are
implied rather than stated outright.
Finally, the Toulmin approach reveals the use of nine Major Claims, all
supported by Major Data and mostly authorized by suppressed Warrants. The key
Major Claim is public safety is among FERCs highest priorities when it comes to
regulating LNG terminals. This claim is supported by Major Data statement FERC
uses a comprehensive siting process (sub-data are provided) and the authorization in
this case depends on the notions that Public safety is important, Government
Agencies exist to ensure public safety, and Ensuring public safety is the job of the
government (all these Warrants are substantive and are all suppressed). This Major
Claim is restated several times in the text, for instance as FERC has been and will
continue to be proactive in addressing safety concerns [MC] because FERC
ensures the safety and security of LNG terminal projects from the time they are
conceived [MD, subdata provided], with the connection between the two
depending on the notion that FERC is a leader [Authoritative, suppressed].

Another important claim made by the text is the market ultimately determines
whether an approved LNG terminal is ever built [MC], because even if LNG
terminal project receives all of the federal and state approvals, it still must meet
complicated global issues surrounding financing, gas supply, and market conditions
[MD, subdata provided], connected by the notions that Market forces control
development and Market forces determine energy policy [Substantive,
suppressed]. Yet another series of three major claims deal with U.S. need for natural
gas, such as The demand for natural gas in the U.S. has been exceeding supply
[MC], because natural gas usage is increasing while U.S. production is falling the
logical movement here dependent on the notion that LNG is the key supply source
[Authoritative, stated]. Another key statement is There is a growing controversy
regarding the safety and security of...LNG ships and terminals [MC], because As
the number of proposals to site LNG terminals in the U.S. has increased, so has the
level of controversy [MD, subdata provided], the movement authorized by the
notions that LNG vessels and terminals are potential terrorist targets
[Authoritative, stated] and that Public perceives LNG as unsafe [Motivational,
The number of key claims shows the complexity of the notions that are dealt
with in the FERC narrative. Most of these notions are centered around the reiteration
of FERCs commitment to public safety, the role of the market in LNG growth, and

the fears about LNG technology, with the primary emphasis on FERCs authority
and expertise in dealing with public safety issues. Each major claim made by the text
is supported by major data statements, with subdata provided, giving the text an
authoritative tone. The major claims are centered around issues similar to those
presented in judgment clusters and clarification devices, giving the text a precise and
informative tone and a tight, well-supported structure (as opposed to rambling tone
and loose structure in which major claims are not immediately followed, or not
followed at all, by major data statements). The number of major claims indicates that
FERC has in mind an audience that is overall well-informed about things such as
market forces, but is not very familiar with LNG issues. In addition, FERC, for the
most part, allows the audience to make the logical move between major claims and
major data, thus indicating that the audience is expected to be sympathetic and share
some of the ideological beliefs implied in the text (such as public safety is
important and government agencies exist to ensure public safety). The warrants
provided by FERC reinforce its authority (FERC is a leader on LNG siting issue),
reinforce the position that LNG is needed by the U.S. economy (LNG is a key
supply source), or allow FERC to appear decisive and transparent on public safety
issues (LNG vessels and terminals are potential terrorist targets). All of the stated
warrants are authoritative, that is, they depend on the audience acceptance of
FERCs credibility.

Overall, the analysis of rhetorical devices reveals that this complex and vast
narrative is mainly centered around reiteration of three key themes (LNG safety, the
need for LNG, and FERCs commitment to public safety and leadership in the field
of LNG siting), as well as around providing information that will allow the audience
to accept the claims being made by FERC. The narrative, by requesting the audience
to make factual judgments, by providing clarification devices for each of the major
three themes, and by structuring itself along the lines of major claims dealing with
the three themes, appears to be highly persuasive and successful in its dual goals to
reiterate FERCs authority and to establish FERCs narrative as the dominant
narrative on the subject of LNG. Thus, we can conclude that the narrative is
successful from the Utilitarian point of view (it does what the author intended to do,
that is to inform and to establish its authority) and from the Scientific point of view
(the claims made in the narrative are extensively supported by evidence which drives
the conclusions and the majority but not all of which can be independently
verified). This dominant narrative can be summarized as follows: The U.S. has a
growing need for natural gas. However, natural gas basins are located outside the
U.S. and so there is a need for technologies that allow us to import natural gas. This
technology is liquefaction. There arent any alternatives to this technology, so it is
here to stay. It is a safe technology, one used around the world and, increasingly, in
the U.S. Even though it is a safe technology, FERC and other government agencies
understand the publics concerns, especially concerns that this technology might

become a terrorist target. FERC is a leader on this issue, an agency empowered by
law to regulate this technology, an expert on the issue that can be trusted, and an
agency committed to ensuring public safety and security.
However, there are silences in the narrative, centered around the issue of
public controversy, that are telling and make one question the psychological
effectiveness of the message (addressing fears and having a desired effect, that of
reduced uncertainty, on the public). Roe insists that we must always think about
uncertainty, complexity, and polarization when thinking about narratives.
Complexity is well-documented in the dominant narrative, but polarization and
uncertainty are not. In this case, the dominant narrative hints at polarization, but
attempts to silence it by not defining the terms and issues that make up the backbone
of the controversy and by making authoritative claims about some aspects of the
controversy (e.g. potential terrorist threat) while ignoring the others. Uncertainty is
also silenced, by linking the audience to the studies that model probabilities of
accidents (such as the Sandia Report), rather than presenting the conclusions of such
studies on the site, as well as by reiterating the regulating agencys authority on the
issue. Because of these omissions, the FERC narrative gives an impression that
FERC is a strong proponent of the use of LNG and an advocate for the industry,
which is not necessarily consistent with its purposes as a regulatory agency and can
create an impression that FERC is not unbiased in its dealings with the industry.

In addition, as laypersons, we have fear of new technologies, because we
dont know what the possibilities and probabilities are. We know what the worst case
scenarios are for oil transportation (Exxon Valdez) and for nuclear power
(Chernobyl), but we dont yet know what the worst case scenario could be for an
LNG accident. Because FERC does present four examples of such accidents, the
danger becomes salient in the readers mind, but the audience (or FERC, for that
matter) doesnt know exactly what the chances are of such accidents occurring. As a
result, we have what Roe calls certainty of uncertainty, a condition which leads to
narrative certainty in opposition stories. That is, stakeholders will subscribe to a
given set of uncertainties (LNG is safe because in 33,000 trips and 60 years weve
only had four accidents or LNG is unsafe because we dont know the extent and
nature of damage it is capable of), with the opposing narrative often creating a
crisis scenario or a calamity too frightening to dismiss.89 A look at the opposition
narratives, coming later on in the work, will reveal the presence of the crisis scenario
in some of these texts.
Industry Story: The Sales Pitch
This section presents an analysis of narratives found on three industry
websites: two websites created specifically to promote proposed LNG terminals that
have met with significant public opposition in Fall River, Massachusetts (proposed
89 Roe, 110-114.

by Weavers Cove Energy/Hess LNG)90 and the Long Island Sound, New York
(proposed by Broadwater Energy/TransCanada/Shell)91, and one website for an
existing terminal, Dominion Cove Point in Chesapeake Bay, Maryland92. Each
narrative will be analyzed using the rhetorical criticism approaches outlined above;
in addition, an attempt will be made to compare and contrast the three narratives and
to identify common narrative themes in the course of the analysis.
Weavers Cove
Weavers Cove is a website that promotes the terminal proposed by oil and
gas corporation Amerada Hess Corporation and Poten & Partners energy shipping
and commodity brokerage firm in Weavers Cove, located in Fall River,
Massachusetts. One of the most striking, for the purposes of this project, features of
the website is that it is explicitly presented as a story. The proposal has met with a
great deal of opposition (which will be addressed in the next section) and the
opening statements on the website address this fact, with the highlighted, large-font
text on the front page asking the audience, Have you heard both sides of the story?
and going on to state, You have probably heard a lot about the proposed Weaver's
90 Weavers Cove Energy. Weavers Cove Energy, Weavers Cove Energy, (accessed March 27, 2006).
91 Broadwater. Broadwater Energy LNG Natural Gas Pipeline, Broadwater, (accessed March 27, 2006).
92 Dominion. Dominion Cove Point LNG, LP, Dominion, (accessed March

Cove Energy facility, but much of what has been said by others is misleading or is
simply untrue. Now you have the opportunity to hear the other side of the story. We
have developed this website to serve as an informational resource dedicated to
dispelling the myths and examining the facts about Weaver's Cove. These few
opening lines set out to accomplish quite a few things, namely: a) acknowledging the
existence of opposition and presenting the Weavers Cove narrative as that of the
underdog, by implying that the opposition side of the story has dominated the
public debate and has received more attention (the words other side imply that the
Weavers Cove story has remained hidden, unheard); b) presenting the oppositions
narrative as misleading, while establishing the Weavers Cove narrative as based
on facts (as opposed to myths); c) most importantly, establishing a relationship
between the rhetor and the audience that is personal (the audience is addressed
directly, you) and based on trust (the audience is invited to hear Weavers Cove
side of the story and to examine the facts for itself); and d) positioning the rhetor
as an objective party that presents the facts and invites the audience to make up its
mind about Weavers Cove.
These last two goals fall in line with the assumptions of the judgmental
approach to rhetorical criticism, which examines the judgments the rhetor is asking
(implicitly or explicitly) the audience to make. The opening statement on the
Weavers Cove website is a factual statement that invites the audience to examine
the facts (those already known and those about to be presented on the website) in

order to make a judgment that the things it has previously heard about the proposal
are, in fact, misleading or simply untrue. It is not surprising that the factual
cluster is heavily developed and that the factual statements appear throughout all the
sections of the website. However, the factual cluster on the Weavers Cove site is
markedly different from that on the FERCs website. FERCs factual cluster is
heavily structured around hard scientific facts about LNG, while the factual cluster
on Weavers Cove website presents both hard facts and statements that depend on
the successful establishment of a relationship of trust with the audience for their
The hard facts are typically clustered around making a case for the need for
LNG (LNG makes up 30% of the gas supply in New England on days of peak
demand), and descriptions of the proposed facility, such as location and capacity.
But it is the statements that depend on establishing a relationship of trust with the
audience for their persuasiveness that are the most prominent. These statements are
typically clustered around LNG safety (In the remote event of an LNG spill from
the facility or a tanker, there would be no explosion) and industry safety record
(With more than 45,000 tanker voyages undertaken since LNG was first transported
by sea, no member of the public has been injured or killed as a the result of an
incident or accident involving LNG shipping; there has been an LNG facility
operating without incident on Bay Street in Fall River for more than 30 years;
There have been accidents in LNG facilities around the world... how ever, any

fatalities...have been confined to within the property lines of those facilities. This is
because of careful design, the use of exclusion zones, and the industrys focus on
safety). Other factual statements in this category ask the audience to agree with the
need for LNG (Economic growth requires energy and fossil fuels are our energy
mainstays; natural gas is one of the most abundant fossil fuels in the world.) and its
environmental friendliness (LNG ranks as one of the safest and least
environmentally hazardous energy sources), as well as with statements about the
advantages of the location (the location has the attributes required for development
as an LNG terminal deepwater access...sufficient land area to allow the required
protection for the surrounding community, close to the pipeline grid, and appropriate
(Industrial) zoning).
Overall, the factual cluster is consistent with the goals set out in the opening
statement it presents the facts about LNG, many of which depend on establishing
a relationship of trust with the audience for their persuasiveness. However, Weavers
Cove narrative doesnt address the myths and misinformation about LNG that are
supposedly being presented by the opponents of the terminal. This allows the
narrative to retain a story structure without becoming an argument that engages the
opposition directly, thus giving spotlight and weight to the oppositions claims. In
addition, retaining the story format allows the authors to combine factual and
desirable statements, the combination that, as Hart and Daughton point out, is often

featured in advertisement.93 In this case, the desirable cluster is heavily developed
(21 statements to 53 factual statements) and mostly concentrates on values of
practicality (An affordable source of gas supply is essential to the reliability of our
electricity supply; In response to growing concerns about the quality of our
environment...power plants in New England have been built with natural gas as their
primary fuel source, as well as statements about natural gas uses in public transport
vehicles in Europe) and beauty (statements about the visual and noise impacts of the
facility and impacts on historic properties around the site, such as Weavers Cove
Energy will consider mitigating visual impacts to the extent improving
the landscaping and vegetation of the site). Interestingly, factual statements are
often followed immediately by desirable statements, inviting the audience to
simultaneously judge the truthfulness of the claim and to consider whether it presents
a desirable state of affairs. For example, the factual statement In Fall River,
residential customers have seen their natural gas bills increase by 50% is
immediately followed by the desirable statement As the demand for natural gas
continues to grow, and if the available supply of natural gas does not grow with it,
further increases in price will almost certainly occur. It might even become difficult
to meet existing residential, business, and power plant requirements.
93 Hart and Daughton, 69

According to Hart and Daughton, in advertising, factual and desirable
statements are often used in combination with directive statements; it is the
frequency, explicitness, and detail of the verbal promises made that allows the
critic to classify the text as a hard-sell or a soft-sell approach. In advertisement,
the combination of factual and desirable statements provides the verbal description
of the context, while directive statements (or photographs) hint at promises for how
to attain the good life.94 In the case of the Weavers Cove narrative, the
combination of the factual and desirable statements describe the energy situation in
New England (with subtle hints indicating the potential for energy crisis), emphasize
the LNG safety record, and present facts about the proposed facility, while asking the
audience to consult its preferences to decide whether or not this context makes it
happy or unhappy in light of values of practicality and beauty, while the directive
statements outline the benefits of adopting a new attitude towards LNG and giving
approval to the facility.
Thus, the first cluster of directive statements, located on the opening sections
of the site, such as Why let others tell you what to think or how to feel about
Weavers Cove? The best way to make an informed decision and figure out how you
really feel is to examine the facts for yourself and We encourage you to read these
statements very carefully and then consider what you may have already heard about
Weavers Cove. The facts may surprise you, explicitly outlines both the course of
94 Ibid.

action (listening to Weavers Cove side of the story and using the website as the
source of information about LNG) and implies the positive consequences of adopting
this course (making up ones own mind, gathering knowledge, and not being mislead
by myths and misinformation into rejecting safe and beneficial technology). In this
case, although these statements are not frequently repeated throughout the text, they
are both bold and unmistakable and fairly sketchy,95 (for instance, myths and
misleading information are not described) and, in combination with their
prominent display on the front page of the site and direct appeal to the audience, give
an appearance of a hard sell of the idea that the Weavers Cove narrative is the one
that contains the facts.
One could argue that the goal of this hard sell is to position the Weavers
Cove story as the dominant narrative on the issue of siting of the proposed terminal
in Weavers Cove. That is, the website first and foremost tries to sell the idea that it
tells the story about LNG siting in the area; if that idea is accepted by the audience,
the narrative doesnt have to make explicit requests of the audience to approve the
Weavers Cove proposal. Indeed, while there are several paragraphs that can be
described as directive, as they outline the benefits of the facility, they do not
explicitly ask the audience to approve the facility, but rather implicitly suggest the
ways in which approving the facility can benefit the region (e.g. Weavers Cove
95 Ibid.

facility will help the local economy. It will mean 30 permanent jobs and as many of
350 jobs during the construction phase. It will also generate $3 million a year in tax
revenues for Fall River; There is an increasing concern at the local and national
level regarding our regions and our countrys ability to meet the growing demand
for natural gas...The Weavers Cove Energy LNG terminal will address this concern
by bringing additional supplies of natural gas to Fall River; as well as several
stories within a story extended paragraphs that present evidence of the growing
need for natural gas that can be solved by increasing LNG capacity, and several
paragraphs that describe the history of the proposed site and the benefits of
converting the existing brownfield to a cleaner, more visually pleasing, more
environmentally friendly Weavers Cove facility). The soft-sell approach to the
directive statements about the facility results in the narrative tone that is not pushy -
Weavers Cove Energy is courting the support of the community, but not in the form
of an aggressive, obvious sales pitch.
The conclusion finds support in further analysis, that uncovers the absence of
desirable/directive statements (used in hard-sell approaches), but extensive use of
factual/directive statements that arent explicit, but allow members of the audience to
come to conclusion that, because good things have resulted from a proposed course
of action in one case, the action would result in good things happening in their own
lives. Thus, the goal of the paragraph that describes the industrial histoiy of Fall
River that features the statement, The city has a rich industrial past, and was one of

the great textile manufacturing centers in the nineteenth and early twentieth
centuries. Immigrant population streamed into the city to work in the booming textile
industry, creating a rich diversity of cultures, is to encourage the audience to
conclude that, since industrial development and new technologies have improved the
economic and cultural fortunes of Fall River in the past, renewed investment in
industrial development can do so again. Similarly, the goal of the section that details
the benefits that Weavers Cove can bring to the city ($3 million in real estate tax,
350 local construction jobs, 30-35 permanent jobs, $4 million worth in purchased
local goods, and $3 million in marine services) is to encourage individual audience
members to conclude that the benefits to the city will be transferable to their own
Examination of Clarification Devices presents a picture that is similar to that
resulting from analysis of the FERC narrative. The narrative is well substantiated by
facts about LNG safety, technology, and benefits LNG presents to the economy
overall and to New England economy in particular. Information about the proposed
facility, pipelines, proposed site, and anticipated economic development benefits to
the community is detailed. Overall, over 70 pieces of information percentages,
dates, numbers, and graphs are presented to support the argument that New
England has a need for natural gas (LNG makes up 30% of the gas supply in New
England on days of peak demand), that LNG is already widely used in New
England (about one-third of the countrys LNG storage tanks are right here in New

England), that the facility can be operated safely and bring economic benefits to the
area, and that the site chosen is appropriate because of its history as an industrial site.
Interestingly, the site makes extensive use of contrast between natural gas
and oil, making an argument that natural gas, by contrast with oil, is more efficient,
clean burning, does not react with water, will not pollute beaches, and will not
necessitate cleanup in case of a spill. Similarly, LNG vessel design is favorably
compared to the oil tanker design (LNG tankers are less vulnerable than crude-oil
vessels because of the ships double-hull construction and separate storage-tank
design; attack on a gasoline tanker could produce an explosion, fire and
environmental attack on a LNG tanker probably would produce a
shipboard fire...with no explosion). The authors in this case are addressing fears
that might result from a potential association of LNG and oil transportation, where
the audience could hold previously formed negative opinions about the oil industry,
based on common knowledge about environmental and safety hazards of oil spills,
and transfer these fears to the LNG industry.
Finally, the Toulmin analysis reveals nine major claims, most centered on
LNG safety and economic need for natural gas. Thus, the movement between the
statement The US gas market has experienced a growing imbalance between
demand and supply [MC], and the statement New England has experienced some
veiy dramatic gas price increases [MD, subdata provided] depends on the
understanding that gas shortages are bad for the economy and consumers

[motivational warrant, suppressed]. The movement between the statement LNG
ranks as one of the safest and least environmentally hazardous energy sources [MC]
and the statement LNG is far less flammable and will not pollute the environment
if it were to spill [D, subdata provided] depends on the notion that we need sources
of energy that are environmentally friendly [motivational warrant, suppressed]. The
movement between statement the best way to make an informed decision and figure
out how you really feel is to examine the facts for yourself [MC] and statements
about LNG safety [MD, subdata provided] is authorized by commonly held notion
that our feelings are shaped by what we know [Substantive warrant, suppressed].
The number of major claims indicates the complexity of the subject and the fact that
most of them are well substantiated by data gives the narrative a well-structured
tone, where a major claim would typically appear at the top of each section, followed
by several pieces of data. Following the soft-sell approach, all of the warrants are
suppressed and all but three are motivational, thus allowing the audience to authorize
the movements between major claims and major data based on its feelings and
desires. The fact that two major claims (A lot of misinformation regarding LNG has
been put forth and this is a place where both sides of the story can respond and talk
openly) are not substantiated by major data, further adds to the impression of an
advertisement, where a desirable condition might not necessarily be supported by

Overall, analysis of rhetorical devices reveals that this narrative is mainly
centered around the themes of safety, New England-specific economic need (unlike
FERCs narrative, which addresses overall U.S need), and the appropriateness of the
selected site. By combining factual, desirable, and directive statements, Weavers
Cove Energy creates a narrative that can be described as a soft-sell sales pitch the
success of which depends on establishing a relationship of trust with the audience.
From the Utilitarian point of view, this is a successful narrative, as it does what it set
out to do presenting the proposed facility in a positive light. Of course, the
narrative omits important facts about LNG-related accidents; this silence makes it
successful from the psychological standpoint (as it calms the fears about LNG) but
unsuccessful from the political standpoint (not only does it not address the goals of
the opposition groups, which will be addressed later, the silence leaves the narrative
open to accusations of dishonesty).
Broadwater Energy
The Broadwater Energy website promotes the proposal by TransCanada
Corporation and Shell to build an off-shore LNG receiving facility about nine miles
from the Long Island Sound (LI Sound) area. Like Weavers Cove, the proposal has
met with significant opposition from local groups and politicians, such as Senator
Charles Schumer (D-NY). Given the wide-spread opposition to the project, it is not
surprising that the Broadwater Energy website features a narrative that attempts to
present a positive image of the proposal and courts public approval, addressing the