Moral Development

MATTHIAS KAISER, KATE MILLAR, ERIK THORSTENSEN, and SANDY TOMKINS

DEVELOPING THE ETHICAL MATRIX AS A DECISION SUPPORT

FRAMEWORK: GM FISH AS A CASE STUDY

(Accepted in revised form October 25, 2006)

ABSTRACT. The Ethical Matrix was developed to help decision-makers explore the ethical issues raised by agri-food biotechnologies. Over the decade since its inception the Ethical Matrix has been used by a number of organizations and the philosophical

basis of the framework has been discussed and analyzed extensively. The role of tools such as the Ethical Matrix in public policy decision-making has received increasing attention. In order to further develop the methodological aspects of the Ethical Matrix method, work was carried out to study the potential role of the Ethical

Matrix as a decision support framework. When considering which frameworks to apply when analyzing the ethical dimensions of the application of agri-food bio- technologies, it is important to clarify the substantive nature of any prospective

framework. In order to further investigate this issue, reflections on the neologism ‘‘ethical soundness’’ of an ethical framework are presented here. This concept is introduced in order to provide more structured evaluations of a range of ethical

tools, including ethical frameworks such as the Ethical Matrix. As well as examining the philosophical dimensions of the method, theoretical analysis and literature studies were combined with stakeholder engagement exercises and consultations in order to review the Ethical Matrix from a user perspective. This work resulted in the

development of an Ethical Matrix Manual, which is intended to act as a guide for potential user groups.

KEY WORDS: biotechnology, decision support, Ethical frameworks, Ethical

Matrix, GM fish

1. INTRODUCTION

One of the first decision-support frameworks to be developed to explore the

ethical issues raised by agri-food biotechnologies was the Ethical Matrix,

proposed by Mepham of the University of Nottingham in the mid 1990s.

Since then, the Matrix has been applied to a number of case studies, such as

bST, xenotransplantation and bioremediation, by Mepham and his col-

leagues at the University of Nottingham (e.g., Mepham, 1996, 2000, 2001;

Mepham and Tomkins, 2003; Moore, 1996; Millar, 2002) and in the Food

Journal of Agricultural and Environmental Ethics (2007) 20:65–80 DOI 10.1007/s10806-006-9023-8 � Springer 2007

Ethics Council, and by others working at several different institutions (e.g.,

Kaiser and Forsberg, 2001; Kaiser, 2004; Schroeder and Palmer, 2003;

FAO/WHO, 2003; Chadwick et al., 2003). Over the decade since its incep-

tion, the Ethical Matrix has attracted considerable attention from specialists

in the field of bioethics, and the philosophical and methodological basis of

the framework has been discussed and analyzed extensively. Therefore, we

shall not attempt to survey and present all pertinent aspects and justifica-

tions of the Ethical Matrix in this paper (refer to Mepham et al., 2006 and

cf. also Forsberg, 2007).

The role of tools such as the Ethical Matrix in public policy decision-

making has received increasing attention over the last few years, with a

number of practitioners exploring the use of new and novel frameworks and

tools, in particular in regard to applied ethics. In order to further develop the

methodological aspects of the Ethical Matrix method, the authors have

studied the potential role of the Ethical Matrix as a decision support frame-

work in the field of biotechnology and food regulation. Theoretical analysis

and literature studies were combined with stakeholder engagement exercises

and consultations in order to study various aspects of the Ethical Matrix

from a user perspective. This work resulted in the development of an Ethical

Matrix Manual (Mepham et al., 2006). The aim of the paper is to clarify some

general evaluation criteria for the uses of the Ethical Matrix as a decision

support framework, and to exemplify briefly two participatory approaches.

2. ETHICAL SOUNDNESS OF THE FRAMEWORKS

When considering which frameworks are appropriate for use when ana-

lyzing the ethical dimensions of agri-food biotechnologies, it is important to

clarify the substantive nature of any prospective framework. In order to

investigate this, we have attempted to present some reflections on the

neologism ‘‘ethical soundness’’ of an ethical framework. This concept is

introduced in order to provide more structured evaluations of a wider range

of ethical frameworks. 1

The term ‘‘soundness’’ in this context is adapted from the philosophical

work on logic. An inference is normally termed ‘‘sound’’ if, and only if, the

logical form of the argument is valid (i.e., truth preserving) and all its

premises are true. An axiom system is sound if, and only if, all the axioms

and theorems are true under all interpretations. Accordingly, one would say

that an ethical argument is sound if, and only if, the logical form of the

normative argument is valid (e.g., in a version of deontic logic, namely, a

1 For a list of several such frameworks see the website of the project Ethical Bio TA Tools:

http://www.ethicaltools.info/.

MATTHIAS KAISER ET AL.66

logic of normative concepts) and all its premises are either true or valid (here

in the sense of the German term ‘‘gültig’’; ‘‘normative rightness’’). It could

then be said that a normative system (theory) is sound if, and only if, all its

axioms and derived theorems are valid under all normative interpretations.

However, an ethical framework is not to be confused with a particular

ethical theory. An ethical theory would, ideally, distinguish all morally right

from all morally wrong or morally neutral actions. A framework on the

other hand only assists us in reaching a reflected ethical insight or normative

conclusion. A framework is not composed of statements that together make

up a normative theory; rather it is a tool on a meta-level, since it makes use

of a variety of ethical theories and normative arguments. Given this variety,

the above description of ethical soundness cannot apply to such a frame-

work, since viewpoints might be included that arise from very different

theoretical conceptions and that may collide with each other. Thus, ethical

soundness of a framework cannot mean normative validity under all

interpretations. But a framework is also a practical and pragmatic tool,

since it should allow us to extract all relevant information for decision-

making involving ethical issues, without necessarily implying a unique an-

swer to all issues. Ethical frameworks are not material objects, or even sets

of statements; rather, they are conceptual or procedural devices, which are

designed to facilitate explicit ethical decision-making, typically by a body

consisting of several individuals with varying viewpoints. They are tools that

are dependent on the competency of users, i.e., as moral individuals. Given

this definition of an ethical framework, the use of ethical soundness in this

context is therefore conjectural and non-traditional, and requires further

analysis. This analysis should take account of the essential function of

ethical frameworks to assist public bodies in making ethically justified

decisions.

The question then arises, what should be preserved and accentuated if

one transposes ethical soundness from normative ethical theory to ethical

frameworks that are developed to aid decision-making. The intuitive notion

is that a decision support framework works well, i.e., is ethically sound, if it

allows competent access to all relevant normative considerations and facts

of an issue, and leaves the users free to draw conclusions based upon their

own priorities, but informed by insights into alternative viewpoints.

Frameworks that have an inbuilt bias towards a particular ethical position

should not be regarded as ‘‘ethically sound.’’ Similarly, frameworks used

within a given knowledge base that systematically neglect information that

is crucial for certain normative viewpoints, are also not ethically sound. All

frameworks should ensure that the normative reasoning put forward by

potential users is transparent to all external reviewers and evaluators.

GM FISH AS A CASE STUDY 67

Therefore it is proposed that an ethical framework is ethically sound, if

and only if, its application produces understanding of ethically relevant con-

siderations in such a way that within a given body of knowledge and on con-

dition of its competent use no further considerations would decisively alter the

normative conclusions drawn from the framework by the users.

Such a definition of soundness of an ethical framework in some way

parallels proposed definitions of, e.g., scientific objectivity (cf. Føllesdal

et al., 1986, pp. 354–360). 2 It is clear that a definition of this kind with an

inbuilt counterfactual clause cannot be checked in any direct manner as to

whether it applies or not. However, a test of ethical soundness is to ask

whether certain ideal criteria are sufficiently respected in a competent use of

a framework, such as the Ethical Matrix. These criteria could then be seen as

indicators of ethical soundness, in much the same way as one talks of

indicators of sustainability etc. The indicators would be such that neglecting

them would weaken the value of the framework as a decision support tool

and typically lead to different normative conclusions or different reasons for

given conclusions. It should also be recognized that these indicators could

be used to evaluate the performance of several alternative ethical frame-

works (of which the Ethical Matrix is just one).

Thus, in the decision-making context addressed here, i.e., a context

where a decision-maker is faced with competing value claims in society and

a plurality of ethical theories appealed to by various groups, several con-

siderations enter the picture that may indicate the ethical soundness of

frameworks. The main properties of ethically sound frameworks, as we

tentatively see them, are the following:

(1) Inclusion of values at stake

(2) Transparency

(3) Multiplicity of viewpoints

(4) Exposition of case-relevant ethically-relevant aspects

(5) Inclusion of ethical arguments

2.1. Inclusion of Values at Stake

Moral decisions represent choices between different moral impacts or

assessments. Therefore, an explicit listing of all the values that are at stake in

a decision is instrumental for the ethical soundness of the framework. If a

relevant value is overlooked, the outcome of the moral evaluation will

typically be different.

2 Føllesdal et al. (1986) characterizes scientific objectivity as the ideal that is reached when a

given scientific claim would not be evaluated differently by adding more pertinent facts in its

justification.

MATTHIAS KAISER ET AL.68

2.2. Transparency

One of the main challenges of ethical frameworks is to overcome the

opaqueness of moral decision-making. The challenge is thus that the process

of decision-making becomes transparent. Transparency is a precondition for

(a) allowing a wide and inclusive social debate on the issue, and (b) making

decisions that are socially robust and not immediately challenged by public

voices on the basis of information gaps. All too often political actors and

decision-makers seem to assume that ethical positions are simply taken,

without clarification of why and how they are reached and how they could

be challenged. This opacity leads to instability in moral evaluation. 3

2.3. Multiplicity of Viewpoints

Democratic societies are by definition pluralist societies, and this extends

naturally to the sphere of ethics. Any given or automatic dominance of a

particular ethical viewpoint, be it a utilitarian viewpoint or Kantian ethics,

etc., would infringe on the rights of those who adhere to alternative ethical

viewpoints. Not seriously considering other ethical viewpoints would also be

deeply unjust and run counter to ideals of equity. An ethical framework

should, therefore, take account of this multiplicity of known ethical view-

points. To the extent that this is not guaranteed by the competent use of a

framework, the ethical soundness of the framework is threatened, since

consideration of alternative conceptions may alter the outcome.

2.4. Exposition of Case-specific Ethically Relevant Aspects

Ethical decision-making presupposes that all ethically relevant aspects of

the issue are adequately accounted for. As well as appropriately specified

ethical principles, ethically relevant aspects include factual information, the

inclusion of which potentially contributes to strengthening or weakening a

particular moral outcome or judgment.

2.5. Inclusion of Ethical Arguments

Frameworks seem to differ in regard to the extent to which they aspire to

represent ethical argumentation. In ethical theory, the ideal is that an ethical

position is reached through a reflective argument, based on the available

3 Even transparent decisions may of course be challenged, but then disagreements may either

point to a lack of specific information that was not duly considered, or to a simple disagreement

on the weighing of values. Currently we see that a number of decisions are challenged simply on

the basis of being ‘‘unethical’’ or for presenting the wrong ethical position. Ethical frameworks

aspire to extend beyond this rhetoric by ensuring the transparency of the ethical reasoning

behind any given decision.

GM FISH AS A CASE STUDY 69

information and the ethical principles of the theory. This applies in par-

ticular to situations where different principles run into conflict and one has

to weigh and balance them. Knowledge of the arguments behind particular

decisions enables rational critique and debate. Thus, we might see this

aspect as instrumental for the ethical soundness of a framework.

The concept of ethical soundness was applied to evaluate the suitability

of different candidates to operate as ethical tools in democratic decision-

making processes. For instance, a committee approach (that might be

ranked as a procedural decision support framework) might not easily, or

necessarily, satisfy all requirements for transparency, e.g., in terms of how

the final recommendation was reached and who discussed what. However, it

is clearly very difficult to provide a definite characterization of all possible

frameworks.

Examining the Ethical Matrix, and the process for translating the prin-

ciples for such a matrix, reveals that the criterion of transparency seems

optimally to be realized. A judgment formed on the basis of a Matrix shows

clearly what information was available and which considerations were

judged to be dominant and how they were weighed. Similarly, one can say

that the multiplicity of viewpoints is optimally accounted for by the explicit

listing of stakeholder viewpoints. In principle, there is also a good chance

that an Ethical Matrix will account for all ethically relevant information and

all ethical arguments. Yet, there are some caveats: these features are

dependent on the information that is available and included in the analysis.

Typically, this will depend on the scientific training and ethical competence

of the users of the framework, such as organizers of a workshop and the

participants or a committee secretariat. To the extent that all users are

challenged by information overload, it may be difficult to be at the cutting

edge of all issues. This may be reflected, for example, in the quality of the

information included in the Matrix. In spite of this caveat, it seems that the

probability of including all relevant values at stake would be high, since this

is a criterion for compiling a list of interest groups. In summary, the Ethical

Matrix would appear to score relatively highly on the indicators of ethical

soundness.

One of the purposes of conducting an exercise to determine ethical

soundness is to recognize that not all frameworks score alike under the five

specified criteria. It appears that no framework could easily fulfill all the

ethical soundness criteria. This may be due to the following factors: (i) the

list of conditions may not really be comprehensive for this purpose; (ii) some

frameworks may consciously be built upon leaving out certain aspects in

order to optimize other aspects, e.g., playing down ethical argument in favor

of transparency and multiplicity of viewpoints; (iii) the nature of ethical

decision-making may be such that one in effect has to choose between a

MATTHIAS KAISER ET AL.70

broad and inclusive framework and an in-depth framework with detailed

expert arguments; and (iv) different frameworks may be intended for dif-

ferent stages of the decision making process. In view of such limitations, one

should be careful to consider the five properties as tentative indicators for

reviewing the ethical soundness of a framework, and not as conditions that

together define ethical soundness.

The aim of the above characterization was to clarify further the needs of

decision-makers with regards to ethical frameworks. While one may easily

agree that an ethical framework should facilitate the decision-making pro-

cess, one also needs to pay closer attention to the variety of possible uses to

which these frameworks can be put.

3. DEVELOPING A USER MANUAL FOR THE ETHICAL MATRIX

When considering the conceptual basis of the Ethical Matrix, as articulated

by Mepham et al. (2006), it is at its simplest level a checklist of concerns,

structured around established ethical theory. However, it can also be used as

a means of promoting structured discussion. The interest groups, the

weighting of each cell, and even the appropriateness of the principles may all

be challenged or modified by those using it. At best, it helps those involved

in making a decision to put themselves in the shoes of others. At the very

least, it ensures that more than the usual narrow range of concerns are

raised and considered.

In order to review the needs of potential users of the Ethical Matrix and

to develop a user manual for the Ethical Matrix, two workshops were

convened, in each of which the potential use of GM fish was chosen as the

case study. This case was chosen in order that the exercises could relate

directly to the decision-making process involving the Norwegian Ethical

Board on the Ethics of Patents. However, it is not the intention of this

article to present a detailed ethical discussion of the chosen case study. The

main focus of this article is to discuss a number of structural features of the

Ethical Matrix approach that are of interest when evaluating it as a decision

support framework. One workshop was conducted in Edinburgh with a

group of experts, while a second workshop was conducted in Oslo with

‘‘lay’’ participants. One of the previously reported strengths of the Ethical

Matrix is that it can be used by numerous groups and does not presuppose

that participants should all be knowledgeable stakeholders.

The research team reviewed aspects of the operation of an expert group

with a group of lay people when using the Ethical Matrix. However, within

a European context there are notable differences in scientific and political

culture. While some countries seek ethical advice from broadly composed

GM FISH AS A CASE STUDY 71

and highly qualified expert groups (which can include lay members), other

countries, notably in Northern Europe, seek to further include members of

the lay public, in the processes that lead to such advice. The research team

worked on the assumption that there may not be any general or a priori

arguments for or against either of these approaches, but that both require

transparent justification of their ethical assessments.

Building on published material on the Ethical Matrix and the practical

experiences of the research team, a provisional protocol was devised as a

basis for the development of a Manual for the framework (see the diagram

in Figure 1 for a summary). This protocol set out a clear methodology for

potential users. The protocol gave guidance on the application of the

method and both Research Groups used this protocol in two workshop

trials. The protocol was then evaluated to determine its applicability and

was further developed. In order to explore the two methodological ap-

proaches to the Ethical Matrix that have emerged, the two research groups

independently applied a top-down approach (TDA), or the classical form of

the method, and a bottom up approach (BUA) that are described below.

The Ethical Matrix may be used in several ways and by different groups of

people, or even by individuals. So organizational requirements are likely to

be quite different in different circumstances, and to be critically influenced by

factors that are not directly related to the nature of the Ethical Matrix as an

ethical tool. Such factors not only include financial and time limitations but

also the degree to which participants in a group exercise are encouraged to set

the agenda themselves rather than follow a prescribed step-by-step proce-

dure. With reference to the latter point, the different ways in which the

Ethical Matrix may be used will be greatly influenced by whether its use

conforms more to a ‘‘top-down’’ approach than to a ‘‘bottom-up’’ approach.

4. ETHICAL MATRIX TOP DOWN APPROACH (TDA)

In a TDA, the specifications of the Ethical Matrix principles are largely set

by the organizers of the workshop, who have acknowledged expertise in

facilitating bioethical deliberation, and play a prominent role in structuring

the exercise.

In order to examine whether expert participants found the TDA Ethical

Matrix approach useful when applied in a participatory setting, a workshop

was organized in September 2005 in the UK. Using the potential use of GM

salmon in aquaculture as a case study, nine ‘‘expert’’ participants applied the

Ethical Matrix to discuss key issues raised by the use of the GM technology.

The notion of ‘‘expert status’’ was designated to the participants on the basis

that they are actively involved with the aquaculture industry, or biotech-

MATTHIAS KAISER ET AL.72

nology research and development. The group�s expertise covered areas such as marine biology; aquaculture industry; fish welfare; regulation; and

molecular biology, etc. The final selection of participating experts was done

by the workshop organizers. The final list of participants was determined by

a combination of considerations, such as interest and willingness to partic-

ipate, area of expertise, recognized standing in the profession, etc.

DEFINE SUBJECT AREA * Consideration of the technological claims

* Consideration of key ethical issues associated with implementing the technology

DEFINE THE ETHICAL MATRIX * Define the relevant interest groups

* Define the specifications of principles for each of the cells

WORKSHOP OBJECTIVES * Define the aim of the workshop

* Define the scope of the discussion * Define the outcomes expected from the workshop

WORKSHOP MATERIALS * Prepare an introductory presentation

* Define the meeting sessions * Define the role of the facilitator

* Prepare a briefing paper on the key ethical issues

FEEDBACK FORMS * Prepare meeting feedback forms to review process and

the methodology * Prepare SWOT analysis forms for assessing the

methodology

SELECT WORKSHOP PARTICIPANTS * Define the relevant stakeholders

* Determine the relevant areas of expertise * Ensure that there is a diversity of opinion and interests

represented

PARTICIPANT INVITATION * Clarify the terms of reference for the workshop

* Clarify the reporting procedure and the role of each participant * Send out participant documents including a description of

the method and topic briefing paper

WORKSHOP * Opening session that defines the methodology, sets out the aims of the meeting, clarifies the context of the discussions and clarifies the reporting process for presenting the results of the

meeting. * Balanced presentation of experts’ views

* Work through the cells of the matrix in a series of discussion sessions * Conclude with an overview session that draws out participants’ overview of the issues

* Completion of the feedback forms

ANALYSIS OF THE DATA BY FACILITATOR * Qualitative analysis of the discussion

* Quantitative analysis of data * Qualitative analysis of the feedback forms

* Report on findings

FEEDBACK TO PARTICIPANTS * Send draft of the workshop data analysis back to participants to allow

them to check and review the representation of their input

FINAL ANALYSIS AND REPORT

Figure 1. Summary of a generic protocol of the Ethical Matrix.

GM FISH AS A CASE STUDY 73

Participants were sent briefing documents prior to the meeting, including a

copy of the Ethical Matrix methodology. The workshop was held over one day

andatthestartof theevent,participantsweregivenanintroductiontotheEthical

Matrix and the potential issues raised by the case study. During this process, the

classic TDA Ethical Matrix was used, where the principles are pre-specified for

each of the interest groups (cf. Table 1). One justification for using the ‘‘classic’’

Ethical Matrix approach was that it had proven useful in many earlier settings,

and that its form and content are well documented (cf. Mepham, 2005).

The participants used the Ethical Matrix to map out the key issued

raised. As part of their assessment, they were asked to comment on the use

of the Ethical Matrix and its potential value. When examining participants� views of the Ethical Matrix in relation to the notion of ethical soundness,

many felt that the strengths of the framework were its inclusion of a mul-

tiplicity of viewpoints and the discussion of the key (ethical) arguments.

Participants commented on the value of working through the pre-defined

Ethical Matrix as it facilitated a structured debate amongst the group.

When asked to complete a SWOT-analysis (setting out strengths, weak-

nesses, opportunities, and threats) of the method, common weaknesses were

described as: the lack of time to discuss each issue and the potential limita-

tions of the knowledge of the participants. Participants noted the need to

ensure that a broader range of stakeholders are involved in the discussion,

therefore, the involvement of additional participants with complementary

backgrounds would have been welcomed by the group. These potential

limitations should be further explored within the context of other partici-

patory methods, since many of the limitations identified can relate to these

methods per se. As part of the written feedback, all participants believed the

use of the Ethical Matrix helped the process. Although this was a limited

exercise conducted in a UK setting, the findings from this workshop appeared

to reinforce the perception that expert groups prefer to work with a TDA.

5. ETHICAL MATRIX BOTTOM UP APPROACH (BUA)

In a bottom-up approach, the organizers provide less explicit guidance, and

defer to the majority views of the (usually) lay participants in specifying the

principles and conducting ethical deliberation.

In order to further analyze and develop the BUA Ethical Matrix method,

a workshop was organized in Oslo (November 2004). The use of GM

salmon in aquaculture was again chosen as the case for study. Ten lay

participants were chosen from a sample of 70 (standardized demographic

balancing was applied) self selecting citizens who had responded to an

advertisement placed in a public newspaper. The ten participants were

MATTHIAS KAISER ET AL.74

T a b le

1 .

S p ec ifi ca ti o n o f th e E th ic a l M a tr ix

u se d b y th e E d in b u rg h w o rk sh o p p a rt ic ip a n ts .

E th ic a l M a tr ix

fo r G M

sa lm

o n (G

en er ic

d efi n it io n s o f p ri n ci p le s fo r th e co rr es p o n d in g in te re st

g ro u p s)

W el lb ei n g

A u to n o m y

Ju st ic e

T re a te d sa lm

o n

W el fa re

o f th e o rg a n is m

B eh a v io ra l fr ee d o m

In tr in si c v a lu e

F is h p ro d u ce rs

E ffi ca cy , sa fe ty

a n d re m u n er a ti o n

F re ed o m

to a d o p t

o r n o t a d o p t

F a ir tr ea tm

en t in

tr a d e a n d la w

A ff ec te d ci ti ze n s

(i n cl u d in g co n su m er s)

S a fe ty

a n d q u a li ty

o f li fe

D em

o cr a ti c d ec is io n -m

a k in g

In d iv id u a l a n d re g io n a l ju st ic e

T ec h n o lo g is t a n d fi sh

b re ed er s

C o m m er ci a l v ia b il it y a n d

w o rk in g co n d it io n s

A b il it y to

in n o v a te

E q u it a b le

tr a d in g (m

a rk et )

sy st em

E n v ir o n m en t

P ro te ct io n o f th e en v ir o n m en t

B io d iv er si ty

o f b io ti c

p o p u la ti o n s

S u st a in a b il it y o f th e

en v ir o n m en t

GM FISH AS A CASE STUDY 75

chosen on the basis of criteria such as gender, age, variety in educational

background, and apparent motivation in application. Participants were sent

briefing documents on the potential use of GM salmon and the workshop

was conducted in two sessions on two separate days. On the first day, the

participants were introduced to the case study and during the second day the

group worked through the issues using the Ethical Matrix.

For this BUA approach, the research group applied a BUA Ethical

Matrix based on the four principles – with well-being specified separately as

increased benefits and reduced harm, autonomy, and fairness. The partici-

pants translated these ethical principles into specifications for the interest

groups and as a result of discussions the principle Autonomy was modified

and the term Dignity was subsequently used in the Matrix. The participants

also added additional stakeholders to the original list. Some participants

argued for the inclusion of ‘‘future generations’’ as a stakeholder group, but

it was agreed that these considerations could be included under the Con-

sumer group. Others perceived ‘‘Research and Knowledge Production’’ to

be an important issue. As a result of this discussion, an additional stake-

holder group, ‘‘Research Community,’’ was added to the Matrix. Five

interest groups were, therefore, used in this BUA Matrix. The Ethical Matrix

articulated and agreed to by workshop participants is set out in Table 2.

It should be noted that the Ethical Matrix method was introduced to the

participants at the beginning of the discussion. Although, participants

appreciated the organizers initial presentation of an Ethical Matrix as a

starting point for their discussions, they claimed that if given more time,

they might have proposed further changes, both in the list of ethical prin-

ciples and in the list of stakeholders. Participants stated that the Matrix

applied during the workshop should be seen as a first approximation to the

problem, but they also highlighted that it would not necessarily rank as a

‘‘natural’’ classification scheme. The organizers interpreted this view as

indicating a level of abstraction in the Ethical Matrix method that requires

some familiarity with, and knowledge of, ethical assessment procedures.

In terms of outcomes from the workshop, much discussion centered on

what the principles ‘‘Increased benefits’’ and ‘‘Reduced harm’’ actually re-

ferred to. It was, for instance, not obvious to the participants that ‘‘Food

safety’’ was a correct specification of reduced harm for consumers.

On the second day, the participants discussed the potential ethical

impacts for the defined stakeholders (interest groups). This was referred to

as specifying the consequence matrix. The organizers proposed a completed

consequence matrix, but the participants wished to propose their own

analysis of the issues without assistance.

In terms of feedback from the participants and in view of the notion of

ethical soundness, some individuals were of the opinion that the ethical

MATTHIAS KAISER ET AL.76

T a b le

2 .

S p ec ifi ca ti o n o f th e E th ic a l P ri n ci p le s b y th e O sl o w o rk sh o p p a rt ic ip a n ts .

G M

sa lm

o n (f a st

g ro w th

+ st er il it y )

In cr ea se d b en efi ts

R ed u ce d h a rm

D ig n it y

F a ir n es s

F is h p ro d u ce rs

In co m e a n d w o rk in g

co n d it io n s

D ep en d en ce

o n n a tu ra l

re so u rc es , p la ce

a n d su p p li es

F re ed o m

to ch o se

F a ir tr a d e co n d it io n s

C o n su m er s

(p re se n t a n d fu tu re )

N u tr it io n a l q u a li ty

F o o d sa fe ty

R es p ec t fo r th e

co n su m er

(f o o d la b el in g )

A ff o rd a b le

p ro d u ct s a n d

g en er a l a v a il a b il it y

T re a te d fi sh

R es is ta n ce

to d is ea se s

A n im

a l w el fa re

F re ed o m

to m o v e

R es p ec t fo r n a tu ra l

p ro p er ti es

E n v ir o n m en t

P ro te ct io n

P o ll u ti o n

S u st a in a b il it y o f

b io lo g ic a l d iv er si ty

R eg io n a l su st a in a b il it y

R es ea rc h co m m u n it y

N ew

th em

es

a n d fu n d s

D ep en d en ce

o f in d u st ri a l

fu n d in g

C h o o si n g o n e� s

o w n re se a rc h

‘‘ U n d efi n ed ’’

GM FISH AS A CASE STUDY 77

arguments were not fully clarified. This may be a limitation of the BUA,

where principles are not predefined. Since this workshop was conducted

with lay participants, it was felt that there should be sufficient time for a

learning process to occur, in terms of the process and the subject. Giving

more time for participants to reflect upon their own judgments could im-

prove the process. The evolution of specification of the different cells and

their contents are dependent upon interpretation, and this interpretation

depends on the participants� background and preparation. This may be seen as a limitation for the use of the BUA approach with lay participants.

However, the completed cells create compartmentalized data on the par-

ticipants� views of the issues. The completion of the cells demanded a fo- cused debate over how one perceives a decision to affect the interested

parties. This leads to a debate over values and potential ethical impacts.

In addition to the comments on the methodology, several participants

reported that some of their views changed as a result of conducting the process.

For instance, while they raised ethical concerns with regards to commercial use

of growth enhanced GM salmon, they also felt that the development of sterile

GM salmon might reduce the significance of some of the major concerns, since

it might address current concerns relating to wild stock levels. Thus they found

that they did not fundamentally object to the use of GM technology in animal

production, but options were dependent on the distribution of implied benefits.

According to the participants, many of the advantages demonstrated in

this study were procedural. There is room for further development of the

methodology as a tool relevant for engaging with lay participants. Some of

these developments are interrelated, e.g., time and clarity. It is conceivable

that some more time – or stretching the exercise over a longer period – could

lead to more reflection on ethical arguments.

CONCLUSIONS

Applying both the TDA and BUA approaches to the case of GM salmon pro-

vided valuable insights that allowed further development of the Ethical Manual.

The workshops highlighted the fact that, although the Ethical Matrix does not

emerge as a very simple tool to use as a participatory ethical framework, it does

show its potential to structure ethical concerns under varying conditions.

It is not possible to conduct a direct comparison of the outcomes from the

two workshops, since the participant sample is too small and the method-

ological approach differed, in terms of preconditions and type of participant.

However, even with these varying conditions, the two applications of the

method, TDA and BUA, have allowed the analysis of some crucial indica-

tors relating to the ethical soundness of decision support frameworks.

MATTHIAS KAISER ET AL.78

Examining the discourse from the workshops, the participants were able

to present their ethical reasoning in relation to the technology that allowed

their evaluations to be more transparent, the reference to ethical values and

argument was made explicit. The use of both methodologies, TDA and BUA,

allowed individuals to present varying ethical viewpoints, even to the extent

that they may not have coincided with participants original viewpoints.

Conducting further participatory events would establish how comprehensive

the recorded considered ethical viewpoints were in this study. With regards to

ethically relevant factual information, both groups seemed to endorse the

view that a broad range of expertise should be involved in the preparation of

technology evaluation of this nature. This represents a significant challenge

and responsibility for the organizers of this type of consultation process.

These trial workshops highlighted that there are further opportunities to

explore and develop the Ethical Matrix for use as a participatory tool.

However, some interesting outcomes were observed from these exercises. The

expert group felt that the use of the Ethical Matrix provided a needed

structure for the discussion, allowing all participants to contribute and

participate. It was interesting to note that the lay panel, which at the start of

the process had voiced some skepticism towards biotechnology, ended up

with a more positive evaluation of the potential use of GM salmon by the end

of the process. No inherent technophobia seemed dominant in their evalu-

ation and it seemed as if the close examination of this case study on the basis

of an Ethical Matrix approach provided an overview that allowed partici-

pants to distance themselves from preconceived ideas and judge on the basis

of information and principles that are designed to serve the common good.

Thus, it is hoped that for both applications, lay and expert, that the Ethical

Matrix may actively contribute to what one may call ethical discourses.

ACKNOWLEDGMENTS

This work was conducted as part of the EC funded project under DG Re- search, FP5, Quality of Life program, ‘‘Ethical Bio TA tools,’’ (QLG6- CT-2002-02594). The funding and the inspiring collaboration of all other consortium members are gratefully acknowledged.

REFERENCES

Beauchamp, T. L. and J. F. Childress (2002), Principles of Biomedical Ethics, 5th ed. Oxford: Oxford University Press.

Chadwick, R. S., B. Henson, G. Moseley, M. Koenen, C. Liakopoulos, A. Midden, G. Palou, D. Rechkemmer, D. Schroeder, and A. Wright (2003), Functional Foods, Berlin: Springer Verlag.

GM FISH AS A CASE STUDY 79

FAO/WHO (2003), FAO/WHO expert consultation on the safety assessment of foods derived from genetically modified animals including fish, November 2003, Rome:

FAO/WHO, pp. 17–21. Føllesdal, D., L. Walløe, and J. Elster (1986), Rationale Argumentation, Berlin, New York: W. de Gruyter.

Forsberg, E. M. (2007), ‘‘Value Pluralism and Coherentist Justification of Ethical Advice.’’ Journal of Agricultural and Environmental Ethics, 20. Doi: 10.1007/ S10806-006-9017-16.

Kaiser, M. (2004), ‘‘Xenotransplantation – Ethical Considerations Based on Human and Societal Perspectives,’’ in W. Farstad, Ø. Andresen, O. Nyberg, and B. Christensen (eds.), Acta Veterinaria Scandinavica, Proceedings of the 16th Inter- nordic Symposium of the Nordic Committee for Veterinary Scientific Cooperation on

Animal Organs to Save Human Lives, Suppl. 99, pp. 65–73. Kaiser, M. and E. M. Forsberg (2001), ‘‘Assessing Fisheries – Using an Ethical Matrix in a Participatory Process.’’ Journal of Agricultural and Environmental

Ethics, 14, pp. 192–200. Mepham, B. (1996). ‘‘Ethical Analysis of Food Biotechnologies: An Evaluative Framework,’’ in B. Mepham (ed.), Food Ethics, London: Routledge, pp. 101–119.

Mepham, B. (2000), ‘‘A Framework for the Ethical Analysis of Novel Foods: The Ethical Matrix.’’ Journal of Agricultural and Environmental Ethics, 12, pp. 165–176.

Mepham, B. (2001), ‘‘Novel Foods,’’ in R. Chadwick (ed.), Concise Encyclopaedia of Ethics and New Technologies, San Diego: Academic Press, pp. 300–313.

Mepham, B. (2005), Bioethics: An Introduction for the Biosciences, Oxford: Oxford University Press.

Mepham, B. and S. Tomkins (2003), Ethics and Animal Farming: An Interactive Web

Program. www.ethicalmatrix.net. Mepham, B., M. Kaiser, E. Thorstensen, S. Tomkins, and K. Millar (2006), Ethical Matrix User Manual, The Netherlands: Agricultural Economics Research Institute.

Millar, K. (2002), ‘‘Thinking about Cleaning Up: The Ethics of Bioremediation.’’ Science and Public Affairs, 3, pp. 20–21.

Moore, C. J. (1996), A Bioethical Analysis of Transgenesis in Animals. PhD thesis,

University of Nottingham. Schroeder, D. and C. Palmer (2003), ‘‘Technology Assessment and the Ethical Matrix.’’ Poiesis Praxis, 1, pp. 295–307.

MATTHIAS KAISER and ERIK THORSTENSEN National Committee for Research Ethics in Science and Technology (NENT) P.O. Box 522 Sentrum, Prinsensgate 18 0105, Oslo, Norway E-mail: [email protected]

KATE MILLAR and SANDY TOMKINS Centre for Applied Bioethics, School of Biosciences, University of Nottingham, Sutton Bonington Campus Loughborough, Leics, LE12 5RD, UK E-mails: [email protected]; [email protected]

MATTHIAS KAISER ET AL.80

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.