Issues of preference and judgement

Expert judgement versus public preference

Likewise, a number of empirical studies have found significant perceptual differences between experts and non-experts. Kaplan (in Nasar, 1988) noted that "although experts are invaluable when used appropriately, they are a dubious source of 'objective' judgements about what people care about in the landscape".

The question of whether the assessment of scenic quality should rely on lay-public or design experts has been the subject of much debate (Robinson et al., 1976; Landscape Research Group, 1988; Kaplan and Kaplan, 1982). According to Craik and Zube (1976), landscape quality can be assessed by means of preferences or judgements. Robinson et al. (1976) suggested that it is important to recognise the difference between preference and judgement, and to decide which type of opinion is to be used as the criterion for assessing visual quality. Judgement is defined as a "critical opinion based on an assessment of merit" or against a standard of comparison "whereas preference is an opinion which specifically relates to a personal 'liking' based on experience" (Robinson et al., 1976).

In practice, the lay-public's visual reactions are often ignored by experts during VIA studies, and in the subsequent decision-making process. Decisions on visual impact are made based largely on their own professional judgements (Sheppard, 1989). Some experts believe that only those trained in design, and experienced in VIA, can express judgements of scenic quality, and to interpret the aesthetic values of the society (Robinson et al., 1976; Jacques, 1980); Craik and McKechnie, 1974 cited in Sanoff, 1991).

It is assumed that members of the lay-public can only express aesthetic preferences which are deemed to be idiosyncratic, arbitrary, and not free from emotive and associational influences (Craik, 1972; Robinson et al., 1976). Experts, on the other hand, can achieve high levels of consensus because they are all using essentially the same cognitive frameworks formed during early training (Kaplan and Kaplan, 1982).

Carlson (1982) has suggested that the public "lacks the experience and knowledge required to be fully sensitive to aesthetic quality" - landscape quality, therefore, can only be analysed by experts from an aesthetic viewpoint, based on normative design standard. However, Arthur (1977) has argued that the implied relationship between professional and public standards of scenic quality must be demonstrated, and not assumed.

A number of authors have advised experts to treat the public as consumers of the environment, and to carry out market research on consumers' taste and preferences, as well as to encourage public participation in landscape assessment as in other planning decision-making process (Land Use Consultants, 1991; Penning-Rowsell et al., 1977; Glasson et al., 1994; Dearden, 1981c; Ribe, 1986; Arthur, Daniel and Boster, 1977; Shuttleworth, 1984; Fortlage, 1990). This is not unreasonable considering that it is the public who ultimately experience the developments. Penning-Rowsell et al. (1977) have suggested that the use of ordinary people to evaluate landscape "would appear to be the most valid method of obtaining socially acceptable landscape evaluations". Glasson et al. (1994) noted that despite the centrality of prediction in the VIA process, many assessments under-emphasise this, at the expense of a more descriptive and qualitative approach.

Impact prediction, and the evaluation of the significance of impacts based on the expert's intuitive judgements, constitute a "black box" in VIA studies in that the intuitive process which leads the expert to his/her judgement on visual impact is not only inaccessible but inexplicable. Glasson et al. have argued that intuition, often wrapped up as "expert opinion", cannot provide a firm and defensible foundation to justify impact predictions.

The possible perceptual differences between experts (i.e. the experienced) and lay-public (i.e. the naïve), and the experts' failure to take public opinion into consideration are part of the problem in VIA studies. Another is the definition of the word "significance", used by experts when judging the degree of impact.

In VIA studies, a proposed development is normally assessed in terms of whether it has "major", "some", "minor" or "no" significant impact on existing landscape quality. The word "significance" carries much weight in influencing final decisions made by local authorities, on whether a proposed development should be approved on visual grounds. In spite of its importance, it has no precise definition. According to the Collin's dictionary, it means 'critical', 'meaningful', 'important', 'vital', 'weighty' etc.. These are highly abstract terms, and do not help experts to be more objective in their predictions, nor allow the local authorities to determine the extent of impact in a measurable degree.

Recent guidelines jointly prepared by Institute of Environmental Assessment and The Landscape Institute (1995), recommended that, "In assessing the significance of landscape and visual impacts, reliance should be placed upon common-sense and reasoned judgement.." The difference between impacts which are of "major" or "some" significance is open to subjective interpretations and argument. In the field of science and mathematics, the term 'significant' is taken to mean 'something outside of acceptable limits'. This limit is usually set at +5% or lower. Section ??? describes a way in which the four levels of significance can be used to describe the level of impact quantitatively.

The possibility of bias

The fact that impact assessments are based solely on professional judgement can also be problematic. It raises the question of impartiality. Two factors may influence the design consultants to be less critical, when predicting the potential impact of proposed development. They are:-

• The fact that the consultants are employed by developers to advise, or carry out, VIA studies may make them less inclined to jeopardise the development.
• As the same experts are obligated to suggest mitigative measures, one can reasonably assume that they may regard these measures to be more effective than in actuality - a phenomenon very much like the "cognitive dissonance" effect. According to Festinger (1964, cited by Weichhart, 1982/83), the effect of cognitive dissonance is a post hoc justification of choices made by enhancing the advantages and reducing the disadvantages of the chosen object - in this case, the mitigative measures. Arguably, experts who are involved in the design of the proposed developments, are less likely to make critical assessment of the intrusiveness of their design on existing landscape quality.

The differences in opinions between architects and non-architects have been attributed to the former's professional training based on the teachings of architectural aesthetic theories (Robinson et al., 1976; Kaplan and Kaplan, 1982, Hershberger, 1969). Whilst landscape aesthetic theories are primarily concerned with man's interactions with the natural landscape, architectural aesthetics theories, on the other hand, are concerned with built forms (particularly buildings).

Architectural aesthetics have tended to be influenced by the concepts and theories which have been used in the analysis of art. Under the influence of art, architectural aesthetics have tended to focus on individual buildings and on the normative judgements of the individual designer. Andrew (1994) noted that this has involved identifying design elements of buildings which are thought by design experts to be "aesthetically pleasing, in much the same way that paint built up on canvas can be viewed as pleasing". Lang (1987) observed that architectural theories which consist of design principles, standards, guidelines and manifestos are based on ideological positions of "what the world, good architecture, landscapes and urban designs should be?".

According to von Miess (1990), such ideologies on what is good and bad; right and wrong; desirable and undesirable in architecture are based on "subjective totality" (i.e. entirely phenomenological) and partial selectivity from a small group of men which are practised by the profession. Appleton (1993, p.77) pointed out that "many of the theories of architectural aesthetics have become hallowed as if they embodied established truth when, in fact, they are no more than speculations fortified by tradition". Murphy (1993) highlighted that the wide range of architectural styles which currently influenced the design of buildings appears to the "lay-person that self-indulgent, egotistical games are being played by so-called professionals at the expense of society at large".

The emphasis of architectural theories has been on buildings rather than on buildings in new surroundings, on the subjective professional judgements of design experts, and on their normative approach toward the assessment and design of buildings in the environment. This emphasis has been criticised for being too fragmented and introverted, in that, there is little or no regard, to the social or natural context (Fladmark et al., 1991; Zevi, 1978; Wohlwill 1980). In forwarding his ecological approach to architectural aesthetics, Carlson (1986 - cited by Andrew, 1994) argued that buildings are intrinsically related to people and the culture that use them, and to other buildings and their surrounding settings.

In his book "The Modern Language of Architecture", Zevi (1978) suggested that the design process must be concerned with: ` ...an architecture that is not isolated but can communicate with its external reality'. Zevi's view is consistent with Neutra's (1954) who stressed that the design profession needs to be guided by "tangible observations rather than abstract speculations". Environmental scientists have also criticised architectural aesthetic theory for its lack of external validity (Lang, 1987).

Under the above circumstances, expert judgements of impact can not be completely objective. The above criticisms point to the importance of taking into consideration lay public reactions and to the surrounding setting of the proposed development in visual impact assessment. In other words, any assessment made by designers should be based on ascertainable facts rather than purely on the intuitive judgements of architects. Ascertainable facts are obtained from "tangible observations" of the physical characteristics of landscape and development, and public response.

Despite the limitations of conventional descriptive and analysis approach in the VIA process, it will continue to be used, because assessments based on the expert judgements are easy to use. It must be said that although some experts disdain public values, there are others in the same profession who are just as keen to produce public-sensitive designs (Nasar, 1988).

Although public participation in decision-making may, in theory, be advantageous, it is in reality, often an "unsatisfactory experience" for all concerned (R. Kaplan, 1985). For instance, there is the logistics involved (such as transportation, time and money) in arranging for members of the public to participate in the decision-making process. It would be impossible and impractical, to survey the public's reaction to the potential impact of each and every proposed development. In this respect, a predictive tool is highly desirable to anticipate their reactions. A visual impact model for predicting the potential impact of man-made structures on the Scottish countryside using public preferences as an index of measuring changes to the existing scenic quality has been developed by Tan (1996). This will allow designers or VIA consultants to identify the magnitude and dimensions of perceived change in the scenic quality of the landscape with a development in comparison with the situation without the development, well before any construction takes place.

VIA is, therefore, a current procedure for measuring/predicting visual impact of some (i.e. large scale) developments on some natural environments. It uses limited (quantitative) tools like ZVI and viewpoint analysis. So far these tools have largely been limited to manual and automated viewshed analysis and graphic representations of what the proposed facility will look like from sensitive viewpoints.

It raises questions of preference vs. judgements. It raises, however two other questions:-

• judgements by architects vs. judgements by all affected
• objects in themselves and objects in context. It also raises questions of terminology - significance and impartiality.

The solution towards developing a predictive model lies in two sets of tools: evaluation and simulation.

References

Andrew, C. (1994) An investigation into the aesthetic and psychological effects of the soiling and cleaning of building facades.

Appleton, J. (1993) Landscape and Architecture. In: B. Farmer and H. Louw, (ed.). Comparison ro contemporary architectural thought. London: Routledge, 74-77.

Arthur, L.M. (1977) Predicting scenic beauty of forest environments: some empirical tests. Forest Science, 23, 151-160.

Arthur, L.M., Daniel, T.C. and Boster, R.S. (1977) Scenic assessment: an overview. Landscape Planning, 4, 109-129.

Carlson, 1982

Carlson, A. (1986) Reconstructing the aesthetics of architecture. Journal of Aesthetic Education, vol 20, 4, 21-27.

Collin's dictionary

Craik and McKechnie, 1976. Perception of environmental quality: preferential judgement versus comparative appraisals. Unpublished manuscript. Uni. of California, Berkeley.

Craik and Zube, 1976

Dearden, P. (1981c) Public participation and scenic quality analysis. Landscape Planning, 8, 3-19.

Festinger,L. (1964). Conflict, decision, and dissonance. London: Tavistock.

Fladmark, J.M., Mulvagh, G.Y. and Evans, B.M. (1991) Tomorrow's architectural heritage - landscape and buildings in Scotland. Edinburgh: Mainstream Publishing.

Fortlage, C.A. (1990) Environmental Assessment: a practical guide. England: Gower Publishing Co. Ltd.

Glasson, J., Therivel, R. and Chadwick, A. (1994) Introduction to Environmental Impact Assessment. London: UCL Press Ltd.

Hershberger, R. (1969) A study of meaning and architecture. Unpublished doctoral dissertation, University of Pennsylvannia.

IEA and the Landscape Institute (1995) Guidelines for landscape and visual impact assessment. Londdon: E and FN Spon.

Jacques, D.L. (1980) Landscape Appraisal: The Case for a Subjective Theory. Journal of Environmental Management, 10, 107-113.

Kaplan, S. and Kaplan, R, (1982) Cognition and environment: fuinctioning in an uncertain world. New York: Praeger. Chapter 4.

Kaplan (in Nasar, 1988) Perception and landscape: conception and misconception.

Kaplan, R. (1985) The analysis of perception via preference: a strategy for studying how the environment is experienced. Landscape Planning, 12, 161-176.

Land Use Consultants (1991) Landscape assessment: principles and practice. Countryside Commission for Scotland.

Landscape Research Group, 1988

Lang, J. (1987) Creating architectural theory: the role of behavioural sciences in environmental design. New York: van Rostrand Reinhold.

Murphy, I. (1993) The impact of the environment: the shock of the new. In: B. Farmer and H. Louw, (ed.). Comparison ro contemporary architectural thought. London: Routledge, 259-261.

Nasar, J.L. (1988) Environmental aesthetics: theory, research and applications. Cambridge: Cambridge University Press.

Neutra, R. (1954) Survival through design. New York: Oxford University Press.

Penning-Rowsell et al., 1977

Ribe, R.G. (1986) A test of uniqueness and diversity visual assessment factors using judgement-independent measures. Landscape Research, 11, 13-15.

Robinson, D.G. et al (eds) (1976) Landscape evaluation - the landscape evaluation research project 1970-1975. University of Manchester.

Sanoff, H. (1991) Visual research methods in design. Van Nostrand Reinhold, New York.

Sheppard, S.R.J. (1989) Visual simulation: a users guide for architects, engineers and planners. New York: van Nostrand Reinhold.

Tan, 1996

von Miess, P. (1990) Elements of architecture: from form to place. New York: van Nostrand Reinhold (International).

Weichart, P. (1982/3) Assessment of the natural environment - a determinant of residential preferences. Urban Ecology, 7, 325-343.

Wohlwill, J. 1980

Zevi, B. (1978) Modern language of architecture. Seattle: University of Washington Press.