03.03 Basslink Case Study  

Published in the Australian Journal of Public Administration • 62(1):80–87, March 2003

© National Council of the Institute of Public Administration, Australia 2003.

Published by Blackwell Publishing Limited

Author: Ronlyn Duncan

Centre for Environmental Studies

University of Tasmania

The effect of the translation and deployment of science by development proponents in the impact assessment process is examined. Basslink, the 300 km power cable to transport electricity across Bass Strait, is the case study. Drawing on the sociology of science, this paper analyses one critic’s contestation of the Basslink proponent’s science. It highlights the extent to which impact assessment statements fortify proponents’ knowledge-claims, obscuring from view the conditionality of science, and making these claims resistant to independent critique, analysis and verification. These circumstances have the potential to undermine the effectiveness, equity and transparency of regulatory instruments and enforcement mechanisms that derive from the impact assessment process as well as the regime’s legitimacy.

The effect of the translation and deployment of science by development proponents in the impact assessment process is examined. Basslink, the 300 km power cable to transport electricity across Bass Strait, is used as a case study. With the Basslink project spanning state and federal jurisdictions, in 1999 the Tasmanian, Victorian and Commonwealth governments established an ad hoc combined impact assessment process to determine the technology’s social, environmental, community and economic impacts. Its broader institutional context will not be canvassed, as the intention of this paper is to shed light on the scientific aspect of the impact assessment regime, which is usually taken for granted as authoritative or left unexamined.

The limitations of the impact assessment process are well known. A particular criticism, which is the point of departure for this paper, is that proponents engage consulting firms to prepare impact assessment statements that are inevitably in their favour. This putative conflict of interest is defended on the basis that consulting firms will not risk their reputations by producing compromised reports (Beder 1993).

An underlying epistemological commitment that supports this defence is that the ‘facts’ speak for themselves and consultants are neutral intermediaries (Lidskog 1996; Collingridge and Reeve 1986).

Theoretical Context
This paper applies a constructivist epistemological perspective to the Basslink integrated impact assessment process. From this standpoint, it is assumed that knowledge-claims can be reconfigured both within and across the domains of science and policy by intermediaries that are not neutral. On this basis, the ‘facts’ do not speak for themselves but are socially and politically constrained, negotiated and constructed (Lidskog 1996).

The Basslink case study shows how the translation and deployment of science by proponents and their consultants can create barriers which obstruct the critique and verification of their conclusions; analyses that are not only expected by members of the public and interest groups, but required by government agencies with oversight and regulatory responsibilities. The case study describes how these barriers can be deconstructed, which provides insight into how they are constructed. This task draws on the work of Jasanoff (1987, 1990), which focuses on regulatory science, and of Latour and Woolgar (1979) and Latour (1987), which maps the construction of scientific facts. Together they outline the broader dimensions of science, namely, its rhetorical utility and conditionality.

Methodological Context
The author attended the public hearings in Hobart during October and November of 2001 in Tasmania, witnessed testimonies and conducted unstructured interviews. Of significance is how one critic, the Tasmanian Fishing Industry Council (TFIC),1 contested the science that underpinned the proponent’s claim of ‘no significant impact’ of the cable on the marine environment (DIIAS 2001:10–60).

Looking beyond accusations of the TFIC and Basslink’s consultants that the other’s claims were not ‘scientific’, which is interminably arguable, it is contended that the TFIC deconstructed the proponent’s science. It did this not by presenting counterclaims (Collingridge and Reeve 1986) or highlighting uncertainty (Jasanoff 1990) — the usual strategies — but by reconnecting the proponent’s universalised claims back to their local context. This was the key to the TFIC’s ‘success’ which reveals how proponents’ claims can be constructed, the strength of such constructions, and the effort required to penetrate them.

The Argument
It will be argued, therefore, that the Basslink Draft Integrated Impact Assessment Statement (DIIAS) served to decontextualise the scientific knowledge-claims used by the Basslink proponent in support of the project. As such, the proponent’s claims were fortified and made resistant to independent critique, analysis and verification. Efforts by the TFIC to have the project’s technology changed made visible the proponent’s decontextualisation. The TFIC’s reconnection with its context of the science used by the proponent highlighted the conditionality of these claims to members of the Tasmanian public and government agencies.

The TFIC’s actions, and the submissions it instigated, brought a number of issues under close scrutiny by the Basslink Joint Advisory Panel (JAP). In its preliminary report, the JAP determined that areas of the proponent’s science had limitations, that the proposed technological option was not appropriate, and that an independent investigation was required to clarify issues of concern (which revealed data inaccuracies in the DIIAS) (JAP 2002:248–50).

A focus on how the TFIC contested the science highlights what can get lost in the translation and deployment of science across and beyond the domains of science and policy and, crucially, the barriers erected by these moves. If they are not recognised as such, the gaps and obstructions can have detrimental ramifications for the effectiveness, equity and accountability of the regulatory instruments derived from impact assessment processes as well as the regime’s legitimacy. Although an exploration of solutions is beyond the scope of this paper, it is intended that this analysis could be useful in future public policy reform.

Background
The TFIC argued that the Basslink technology should be changed from a monopole to a bipole configuration (TFIC 2001a). The difference relates to the existence, or not, of a return-cable, and the use of sea electrodes. A monopole system does not have a designated return-cable through which current can flow to complete the electrical circuit. Instead, sea electrodes (with the cathode on the Victorian side and the anode on the Tasmanian side) would be installed several kilometres off-shore and away from the power cable to return the current through the sea and the sea bed.

Impacts relating to the monopole system, which were disclosed in the DIIAS, are threefold: electrochemical reactions take place at the anode whereby chlorine is produced; metallic infrastructure, within a specified radius of the anode and with a given orientation and length, is susceptible to accelerated metallic corrosion; the power cable generates a magnetic field which combines with the earth’s magnetic field (DIIAS 2001). In contention were the magnitude, extent and mitigation costs of these effects. The TFIC claimed that the anode’s chlorine could have detrimental effects on the marine environment, and that the magnetic field could affect the migration of marine species, as well as navigation instruments (TFIC 2001a).

Infrastructure owners were concerned that their ports, pipelines and submarine installations would corrode faster than expected. The more expensive bipole system has a designated cable through which current flows to complete the electrical circuit. The TFIC argued that a bipole system would eliminate the marine impacts (TFIC 2001a). It was claimed by the proponent, however, that it would cost an extra $150–200 million, and this would make the project economically unviable (BPL 2001a).

Rhetorical Utility of Science
In contrast to the idealised science-in/policyout model of regulatory policy development (Jasanoff and Wynne 1998:8) the work of Jasanoff (1987, 1990) illustrates the rhetorical utility of science. Jasanoff demonstrates that the turf staked out as ‘scientific’ becomes sacred ground in the regulatory process, and stakeholders squabble over it. This is evident between the proponent and the TFIC in their submissions BPL 2001b; TFIC 2001a).

Contention occurs because it is assumed that what is designated as 'scientific' has the potential to deliver legitimacy. What is at stake is where the line is to be drawn to designate what is credible and defensible, and what is political, which is considered to be arbitrary and contestable. Jasanoff (1987:199) argues that ‘boundary-defining language’, which designates what is and is not authoritative, is the key to these manoeuvres — it determines where power will be held.

Conditionality of Science
The rhetorical dimension of science obscures its conditionality. In the domain of environmental regulation, where defensibility is a prime objective of regulators and stakeholders, the limited variables and narrow sets of circumstances to which scientific claims apply are usually not made explicit (Wynne 1992). Consequently, when science meets policy, the rhetorical dimension of science screens the contingencies from view.

Wynne illustrates this point: … ignorance is endemic to scientific knowledge, which has to reduce the framework of the known to that which is amenable to its own parochial methods and models. This only becomes a problem when (as is usual) scientific knowledge is misunderstood and is institutionalised in policy making as if this condition did not pervade all competent scientific knowledge. This institutionalised exaggeration of the scope and power of scientific knowledge creates a vacuum in which should exist a vital social discourse about the conditions and boundaries of scientific knowledge in relation to moral and social knowledge (Wynne 1992:115).

Constructing Facts, Obscuring Conditionality
Like Jasanoff, Latour and Woolgar (1979) and Latour (1987) seek to demonstrate the rhetorical dimension of science and show how its conditionality becomes obscured by the translations that occur within the domain of science. In this analysis, these theoretical ideas are extended beyond the institution of science and used to describe translations and deployments that can occur across the spheres of science and policy.

Latour and Woolgar (1979) and Latour (1987) examine what scientists actually do in their laboratories. They argue that ‘facts’ do not start out as such, but as statements (or knowledge-claims) that have the potential to become ‘facts’. ‘Facts’, these authors argue, are products of natural, social, cultural and technological interactions, negotiations and translations. ‘Facticity’ is a ‘collective process’ (Latour 1987:29). By observing science-in-themaking and how knowledge-claims are deployed within the scientific community, Latour and Woolgar (1979) introduce the concept of a process and, consequently, the dimensions of time and distance into the formation of ‘facts’. They are not discovered, these authors argue; ‘facts’ are constructed, and start out as signals (ie ‘artefacts’) and ideas. Becoming ‘reality’ takes time.

Moving Modalities
Latour (1987) describes the ‘fact’-making process as involving statements made by scientists being moved along a continuum polarised between ‘facts’ and ‘artefacts’ with ‘facts’-in-the-process-of-being- constructed in between. This takes place through the processes of interaction, review and publication within the scientific community. Movement involves a statement accruing and shedding positive or negative modalities (Latour 1987). Modalities are a type of marker or qualification. They indicate how or under what circumstances a statement is conceived or constructed. Modalities gauge a statement’s conditionality. For example, reference to the use of an ordinal regression analysis (Fuchs 1992) contributes to pushing a statement made in a scientific paper in a positive direction (Latour and Woolgar 1979).

This movement in the direction of the status of ‘fact’ occurs because an ordinal regression analysis is viewed within the particular scientific community as a legitimate and credible means of analysing results. These analytic methods are described as ‘inscription devices’ because they transform the readings taken from technical apparatus in laboratories or the field into forms of representation that are socially acceptable and decipherable, such as graphs, figures and models (Fuchs 1992:68). If a statement is no longer attached to its modalities, it is no longer connected to its source of production. Under these circumstances, ‘facts’ are difficult to deconstruct, or re-open (Latour 1987).

Constructing ‘Reality’
Modalities denote the credibility, or otherwise, of the methods underlying a statement’s production. They help a statement to be assessed within the scientific community and, thereby, moved towards the status of ‘fact’, or not, as the case may be. Negative modalities result in a statement’s movement back towards the pole of ‘artefact’ if, for instance, a given statistical analysis method is superseded or new methods are agreed to be better representations of‘reality’. A statement becomes a ‘fact’, Latour and Woolgar (1979) argue, when it is no longer connected to the modalities about how, when or where it was produced. It no longer needs them. Retrospectively, it appears universal and always to have existed (Fuchs 1992). It has become ‘reality’.

Constructing Barriers
It is the rhetorical dimension of science and how modalities, or ‘traces of production’, are obscured, revealed and made resistant to deconstruction that are of consequence for this analysis of Basslink (Latour and Woolgar 1979:176). With those of Jasanoff (1987, 1990), these ideas provide a framework to analyse the role of the DIIAS, and the actions of the TFIC to draw into question the proponent’s claim of ‘no significant impact’ from Basslink in Bass Strait (DIIAS 2001:10–60).

Not including maps, figures and appendices, the DIIAS contains over 5,500 pages of text collated into chapters, supporting studies and supplementary reports. If stacked, it would stand over a metre high. The text makes reference to hundreds of scientific papers and reports. The pages display innumerable graphs, tables, diagrams and pictures. These forms of representation depict a social, natural and economic ‘reality’ that is the product of the inscription devices described by Latour and Woolgar (1979) and Latour (1987) and project the rhetorical dimension of science, namely, objectivity, defensibility and credibility. It gives the impression that the ‘facts’ do speak for themselves.

Reconnecting Modalities
Despite the proponent’s claims of ‘no significant impact’ on the marine environment from Basslink, the TFIC executive officer (EO) was not satisfied that this finding could be drawn from the information contained in the DIIAS. A particular concern was that the supporting studies related to the marine environment drew on scientific research carried out in the northern hemisphere, from the Baltic Sea, where the salinity conditions and species composition are different from that of Bass Strait (TFIC 2001b).

What were presented as ‘facts’ in the DIIAS were recategorised by the TFIC EO as statements connected to modalities anchored in the northern hemisphere. Crucial to the argument being advanced of a reconnection of the proponent’s claims to their context is that the TFIC sent its EO to Sweden and Denmark, where a number of studies had originated. Over four weeks he met with many experts. With what follows, a portion of the information gathered, the TFIC EO reconnected the statements, presented as authoritative and ‘factual’ in the DIIAS, back to their source of production. The details were presented by the TFIC at a public forum in Hobart on 21 August 2001.

Reconnecting Context and Demonstrating Conditionality
Following a recommendation from the proponent’s consultant, the TFIC EO met with the senior scientist of the Swedish National Board of Fisheries, the lead author of Westerberg and Begout-Anras (1999). The findings of this published research are used in the DIIAS to support the claim that it was expected that there would be ‘no significant impact’ of magnetic fields from the cable on the migration of bony fish and eels (DIIAS 2001:10–60–2). A comparison of Westerberg and Begout-Anras (1999) and the DIIAS illuminates the conditionalities that concerned the TFIC EO.

In Chapter 10 of the DIIAS the following statement is made: The National Board of Fisheries … of Sweden commissioned field experiments to test whether the operating HVDC [high voltage direct current] cable was an obstacle to eel migration… The results from this study showed that all westerly swimming eels crossed the HVDC cable with some showing minor deviations (offsets) in their line of passage (trajectory) … The principle [sic] conclusion reached by Westerberg and Begout-Anras (1999) was that the cable’s magnetic field did not pose an obstacle to successful eel migration … The above overseas studies have indicated that the effects of magnetic fields surrounding these sea cables have little or no effect on Atlantic salmon or eel migration. A similar effect on Pacific eels may be expected, that is a minor deviation when passing over the Basslink HVDC cable but with no significant impact on migration (DIIAS 2001:10–61–3, emphasis added).

In contrast to the above, Westerberg and Begout-Anras (1999) indicate that all ‘westerly swimming eels’ (DIIAS 2001:10–61) did not cross the cable — 12 out of 21 did, and they did so in two different years as shown by the following extracts: To test if the cable was an obstacle for the migration a few eels were tracked in 1997. The results were enigmatic, with two out of four tracks showing very large deviations along the cable in the opposite direction to what should be expected from a simple compass orientation model. A larger effort was made during the 1998 migration (Westerberg and Begout-Anras 1999:150). In all, nine eels were tagged in 1997 and four … passed the cable … In 1998 a total of 17 eels were tagged and of those eight were tracked passing the cable. In total 21 active eels were tracked … and of those 12 (57 percent) passed over the cable (Westerberg and Begout-Anras 1999:152–3).

The discrepancies raised a number of concerns for the TFIC EO. In the first instance, two studies had been combined in the final results, which were undertaken at different times and by different methods. Second, both studies involved very small sample sizes when compared to the size of the fisheries in question (eg thousands of tonnes of eels). Third, it was not appropriate to extrapolate the eel studies to bony fish. Last, the results were seen to present a significant problem for eel migration if only 57 percent crossed the cable (TFIC 2001b). The DIIAS did not show the eel tracks set out in Westerberg and Begout-Anras (1999) from the 1997 study where the eels moved in a predominantly southern direction, parallel to the cable. The tracks of seven eels from the 1998 study were included (i.e. Figure 10.9 of the DIIAS) and illustrated an erratic but eventual western movement of the eels across the cable. In Westerberg and Begout-Anras (1999:157) the difference between the 1997 and 1998 tracks was attributed to the 1997 study being ‘an accidental coincident, without connection with the magnetic field anomalies’.

The TFIC EO described these inconsistencies to the public forum as follows: 'The [eel] study is actually two studies, the first taking place in 1997. The results were not as good as were hoped for, as the eels were not so cooperative as to swim straight over the cable. Nine eels were fitted with transmitters. Four eels crossed the cable, one after demonstrating severe confusion. One also crossed the cable briefly and returned to the eastern side. That was a success rate of only 33 percent … This second study combines the results of the first study and a total of 12 eels from this combined data managed to cross the cable out of 25 eels. The paper from the work (that you have among your notes) explains this as 57 percent. If this is not indicating a problem for a species migrating for reproduction, then what does?' (TFIC 2001b:9).

The TFIC EO had raised his concerns directly with Westerberg in Sweden and reported to the forum that the senior scientist had indicated to him that, when compared with the size of the fishery, the studies were ‘statistically insignificant as well as inconclusive’ (TFIC 2001b:10).2 In addition, the TFIC EO pointed out at the forum that Westerberg had indicated to him that there was a possibility that the Baltic cable (which links Sweden and Germany) was not in continuous operation when some of the eels crossed the cable in the 1998 study (TFIC 2001b).3

On making inquiries about catch data from the Baltic eel fishery, the TFIC EO explained at the forum that Westerberg had indicated to him that he had data that showed a ‘significant increase in catch numbers’ (TFIC 2001b:11). The explanation advanced by the scientist for the increase was that the eels, during their annual migration, were encountering the magnetic field from the sea electrode cable, which is a separate part of the cable system. It extends only partially into the sea, and was not the subject of the tracking studies. Westerberg explained to the TFIC EO that when disoriented, the eels turned away, swimming parallel to the cable until encountering the shore. At this point they would swim back in the direction from which they had originally come and would be caught in fixed eel traps. Traps on the other side of the electrode cable had significantly lower catch rates. The magnetic field from the cable was guiding the eels into the traps by way of an almost circular trajectory, which increased the catch numbers (TFIC 2001b).4

Deconstructing Barriers
It is argued that with information such as this, the TFIC EO recontextualised the decontextualised. At the public forum, fishing industry members and affiliates, government representatives, infrastructure owners and interested members of the public were presented with a bundle of documentation that included translated copies of letters, media, meeting notes, and scientific papers used in the DIIAS.

Using its own rhetorical strategies, the TFIC brought into view what it considered were the contingencies of the proponent’s claims. The JAP received 155 Tasmanian submissions (JAP 2002).5 Of these, approximately 120 expressed a concern about the monopole system. References to the TFIC occur throughout. It is possible that the TFIC’s not-so-good environmental record added an element of credibility to its claims. This is alluded to in a public submission which makes reference to the concerns raised by the TFIC, and comments that it ‘is hardly an outspoken conservation organisation’ (T63 2001:2).

Importantly, assisted by the TFIC’s research, state and federal government agencies took up the issue, as did several Bass Strait infrastructure owners. Given the level of concern expressed in the public submissions about the monopole system, which was supplemented by two independent reports (one undertaken before the hearings and one after), the JAP recommended that the technology should be changed. It is argued that what the TFIC did was bring back from the northern hemisphere information that reconnected statements about the monopole technology and the marine environment in the DIIAS to their source of production.

By reconnecting the modalities, the proponent’s claims were opened up and their conditionality revealed to those critical of the project, government agencies with regulatory and oversight responsibilities, infrastructure owners and, in the end, the JAP with its recommendation that the technology should be changed.

Modality-shedding Devices
Using the theoretical context outlined, the Basslink case study highlights how impact assessment statements draw together, and present as authoritative, statements that could be at any location along the ‘fact’–‘artefact’ continuum described by Latour and Woolgar (1979) and Latour (1987). When a proponent’s consultant translates scientific knowledge claims during the preparation of impact assessment documentation, the modalities of the scientific statements are shed. The act of doing this presents scientific statements as ‘factual’ — black boxes — that are difficult to re-open. The error is that publication does not equate with ‘fact’-status. Publication is the process by which modalities are changed and statements moved between the continuum of ‘artefact’ and ‘fact’ within the scientific community.

It is on this basis that it is argued that the Basslink DIIAS and similar impact assessment documents should be conceived as ‘modality-shedding devices’. They draw on scientific knowledge claims, translate them and, unwittingly or not, configure them into ‘fact’-status statements. The consequence is that the conditionalities of scientific knowledge-claims used by proponents can be obscured from view. It has been shown that bringing them back into focus can be particularly difficult. This is demonstrated by the work of the TFIC which cost the organisation in excess of $20,000, including wages for its EO to work on Basslink full-time for more than four months. The TFIC considered this was a high price for an industry association and that it should not have had to do the work of the proponent or the responsible government agencies (TFIC EO, personal communication).

Whether the contingencies foregrounded by the TFIC were right or wrong is endlessly contestable. Looking beyond such technical debates, this paper has sought to describe the strength of the barriers that proponents can erect in their translation and deployment of science. The predicament is that proponents have the upper hand in the impact assessment process as they are able to stake out the ‘scientific’ turf from the outset. This is not to say that opponents would not construct similar barriers if they had the opportunity to translate and deploy science as proponents now do.

Given the existing process, the Basslink case study demonstrates the pivotal role of critique. With heightening complexity of regulatory issues and an impact assessment regime that facilitates the construction of barriers via the translation and deployment of science, one has to ask how regulatory agencies, with limited personnel and stretched budgets, can undertake the level of critique that this case indicates is necessary to deliver the appropriate regulatory outcomes. It is a concern that the role of critique is falling onto interest groups and members of the public. Except under extraordinary circumstances such as those demonstrated by the actions of the TFIC, their input too is constrained by the mechanics of the process and the deployment of science by proponents.

Conclusions
This analysis suggests that not only can proponents within the impact assessment regime present a potentially biased case on behalf of their own development (cf Beder 1993), but beyond that, proponents’ claims can be fortified, making them resistant to critique, analysis and verification. At the point a scientific paper is translated by a consultant, acting on behalf of a development proponent for its incorporation into an impact assessment report, the modalities of the scientific statements are shed. Yet the attribution of ‘fact’-status would be unwarranted and certainly contestable in other arenas. Resistance to deconstruction and a reliance on the notion of ‘fact’-status was demonstrated at the Basslink hearings.

Under cross-examination the proponent’s expert witnesses defended their claims of ‘no significant impact’ by reference to published papers. Conversely, critics’ claims were dismissed on the grounds they were not verified by publication (BPL 2001b, see endnote 4). Indicative of the rhetorical utility of science and a commitment to the notion that the ‘facts’ speak for themselves, no consideration was given to the intervening translation made by the proponent and its consultants; their claims were accepted as ‘scientific’ and, therefore, authoritative. It is well known that the impact assessment process presents a number of challenges to public policy. These can only intensify.

This paper, which has focused on the scientific aspect, has sought to describe the effect of giving proponents the upper hand in the translation and deployment of science. It has been argued that the scrutiny of proponents’ claims by governments, interest groups and members of the public can be seriously inhibited, if not excluded, by the mechanics of these translations and deployments in the impact assessment process.

As such, the process presents at least two barriers to public policy. An obstruction of critique, in the first instance, calls into question the effectiveness, equity and accountability of the regulatory instruments and enforcement mechanisms that derive from the impact assessment process. Second, it hinders the inclusion and contribution of members of the public and interest groups from which the system derives its legitimacy. Although an exploration of possible reforms is beyond the scope of this paper, a suggestion is that, at the very least, the notion that the ‘facts’ speak for themselves could be jettisoned.

Notes

1. References to the Tasmanian Fishing Industry Council (TFIC) are made with its permission.

2. The Basslink proponent countered this claim (BPL 2001b) by arguing that Westerberg was referring to the statistical significance of magnetic compass orientation exhibited by the eels. It was explained to the JAP that no statistical analysis, only percentages, were required to assess the migration over the cable. The proponent’s expert argued that the percentage of eels that crossed the cable (ie 57 percent) is what would have been expected with or without tracking.

3. The proponent countered this claim to the JAP (BPL 2001b) by indicating that the eel studies commenced at sunset, which would have been during a peak time for electricity use in Sweden. Therefore, it is claimed that the cable had to be in operation. Results tabulated in Table 3 of Westerberg and Begout-Anras (1999) indicate an electrical current reading of zero for two eels that crossed the cable.

4. The proponent’s expert countered this claim to the JAP by arguing that Westerberg’s scenario and diagram was: ‘conjecture and what might happen if the electrode cable was an obstacle … This study referred to by TFIC needs to be published or the data presented to confirm the significance of the higher eel catches in these two traps’ (BPL 2001b:36).

5. Submissions for Victoria totalled 381 (JAP 2002). The health and visual effects of electricity pylons and overland transmission were of particular concern to the people of Victoria.

References

Beder, S 1993 ‘Bias and Credibility in Environmental Impact Assessment’, Chain Reaction 68:28–30.

BPL (Basslink Pty Limited) 2001a Final Submission of Basslink Pty Limited, TE150, Resource Planning and Development Commission, Hobart.

BPL (Basslink Pty Limited) 2001b Evidence to JAP: Response to TFIC Submission, TE40, Resource Planning and Development Commission, Hobart.

Collingridge, D & C Reeve 1986 Science Speaks to Power: The Role of Experts in Policy Making, Frances Pinter (Publishers), London.

DIIAS (Draft Integrated Impact Assessment Statement) 2001 Available from Basslink Pty Limited, PO Box 6141, St Kilda Road Central, Melbourne, VIC 3004 or at <http://www.basslink.com.au/ nw3. html>.

Fuchs, S 1992 The Professional Quest for Truth: A Social Theory of Science and Knowledge, State University of New York Press, Albany.

JAP (Basslink Joint Advisory Panel) 2002 Basslink Proposed Interconnector Linking the Tasmanian and Victorian Electricity Grids: Draft Panel Report, March, Resource Planning and Development Commission, Hobart.

Jasanoff, S 1987 ‘Contested Boundaries in Policyrelevant Science’, Social Studies of Science 17:195– 230.

Jasanoff, S 1990 The Fifth Branch: Science Advisers as Policymakers, Harvard University Press, Cambridge, MA.

Jasanoff, S & B Wynne 1998 ‘Science and Decisionmaking’ in S Rayner and EL Malone eds Human Choice and Climate Change, Volume One: The Societal Framework, Battelle Press, Ohio.

Latour, B 1987 Science in Action: How to Follow Scientists and Engineers Through Society, Harvard University Press, Cambridge, MA.

Latour, B & S Woolgar 1979 Laboratory Life: The Social Construction of Scientific Facts, Sage Publications, Beverly Hills.

Lidskog, R 1996 ‘In Science We Trust? On the Relation Between Scientific Knowledge, Risk Consciousness and Public Trust’, Acta Sociologica 39:31– 56.

TFIC (Tasmanian Fishing Industry Council) 2001a Submission to Joint Advisory Panel, T118, Resource Planning and Development Commission, Hobart.

TFIC (Tasmanian Fishing Industry Council) 2001b Basslink Forum Presentation Notes, Tasmanian Fishing Industry Council, Hobart.

T63 2001 Public Submission to JAP dated 29 August 2001, Resource Planning and Development Commission, Hobart.

Westerberg, H & ML Begout-Anras 1999 ‘Orientation of Silver Eel (Anguilla anguilla) in a Disturbed Geomagnetic Field’ in A Moore and I Russel eds Proceedings of the Third Conference on Fish Telemetry in Europe, Norwich, England, 20–25 June, CEFAS, Lowestoft.

Wynne, B 1992 ‘Uncertainty and Environmental Learning: Reconceiving Science and Policy in the Preventive Paradigm’, Global Environmental Change June:111–27.