Future Comforts workshop

The day of the Future Comforts workshop began with snow, ice and general disruption. The breakdown of the Policy Studies Institute heating system meant that the 17 thermal comfort experts who took part in this event experienced indoor temperatures of around 15 degrees centigrade for the whole day. This provided a tangible reminder of the fact that buildings sometimes fail to deliver what their users expect by way of comfort. In such circumstances, making oneself ‘comfortable’ becomes a matter of individual ingenuity and adjustment. Only a couple of participants challenged social conventions so far as to put their (outdoor) coats back on, but it is probably true that those equipped with woolly jumpers and fleeces fared better than those without.

Setting more immediate concerns about our own comfort aside, we began by confronting a paradox of a much greater order and magnitude: the very systems that are designed to provide comfort indoors are contributing to global warming. Developed in response to such concerns, the Future Comforts project looks at how future concepts of comfort are being constructed today and with what implications for indoor conditions and associated patterns of energy demand.

The purpose of the workshop was to open up debate about what counts as comfort and how it might be achieved under changing (outdoor) climatic conditions - including global warming. The event provided an opportunity for engineers, designers, architects and manufacturers to reflect on the kind of indoor environments they are specifying and constructing and on the global environmental implications of current practice. Participants were invited to think about how concepts of comfort have changed over time, and to consider what new definitions and understandings might emerge in the future.

This report is in three parts. The first summarises and comments on the opening presentations. The second takes stock of insights and observations arising from an afternoon of short, sharp interventions. Part three reviews the results of the second day, a day in which a smaller group met to reflect on emerging themes and key questions for research and policy.

The programme for the first day began with a review of the Future Comforts project and introduced a framework with which to compare different theories of comfort and the methods and approaches they support.

The Future Comforts project has involved a selective review of literature on thermal comfort and interviews with people specifying, designing and producing indoor environments in the UK. The results of this research have been used to develop and refine a typology of comfort paradigms. [1] This simple scheme isolates three contrasting interpretations. Although we might debate the extent of slippage between these three positions, each generates significantly different conclusions for future policy and practice. If comfort is considered a physiologically determined state of affairs, the environmental challenge is to maintain standardised indoor conditions using less resource intensive means. By contrast, defining comfort as a continual achievement supports the development of strategies that maximise opportunities for environmental modification and localised adjustment. For those taking comfort to be a socially and culturally malleable concept, it is possible to conceive of more radical possibilities including the reinvention of comfort and the development of comfort-related expectations that are much less demanding to meet.

Which of these three positions dominates (in policy and in practice) will have implications for the future construction and specification of indoor environments, for the overall direction of social and technical change, and for environmental sustainability.

Having introduced the issues we hoped to address in the workshop, the remainder of the morning session featured papers from three speakers each selected to illustrate different understandings of comfort in particular climatic or cultural contexts. Our first speaker, Koen Steemers, emphasised the diversity of thermal experiences and opportunities encountered in the urban (outdoor) environment.

Koen’s paper raises two important issues concerning the definition and evaluation of comfort in urban environments. One is that conventional methods for assessing comfort need to be modified if they are to be used in evaluating the wide range of micro-climatic conditions that make up the urban outdoor environment. This is partly because people have quite different expectations and perceptions of what it is to be comfortable outdoors. A second point concerns the extent to which achieving comfort in cities is a matter of mitigating microclimatic effects or of maintaining a landscape of thermal diversity.

Although phenomena such as the ‘heat island’ effect are assumed to have negative consequences for the comfort of urban dwellers, Koen suggests that this may be exaggerated. High urban densities and narrow streets generate environments in which people find shade and shelter. The argument presented is that comfort is a diverse and transitory achievement accomplished not by the provision of standardised conditions but through deliberately fostering localised micro-climatic variation. When out and about in cities, Koen argues that people want the choice of whether to stand in the sun or the shade and of whether to enjoy the breeze or shelter from the wind. He concludes that choice itself has a significant influence on people's perceptions of comfort.

However, the ability to achieve 'comfort' is not simply determined by the physical form of the built environment. It is, for example, relevant to think about how free people are to move in and out of pubic and private spaces within the city. Just as important, property values evidently influence the design of urban spaces. In his talk, Koen showed how urban diversity profiles can be generated to reveal the portfolio of environmental and climatic diversity on offer in different urban areas. A next step might be to map patterns of social and economic inequality in terms of people's ability to access different comfort-making opportunities within these urban environments.

There is little doubt that air-conditioning has played a significant role in transforming the global indoor environment over the last few decades, but the extent and pace of change has varied widely in different countries. The job of our next two speakers was to examine and explain such trends. First, Marsha Ackermann reflected on how air-conditioning became ‘normal’ in the United States and how this might relate to developments in the UK today.

One important lesson that Marsha highlights is that predictions can be very wrong. A study in 1948 estimated that at best 3,000 US homes would install air-conditioning between 1948 and 1960, and that this would be limited to richer households in the South. In reality around 6.5 million US households had air-conditioning by 1960. Why are Americans so keen on air-conditioning, and is this set to become a global ambition?

Echoing some of Koen's conclusions, Marsha suggests that the desire for coolth may relate more to issues of ‘climate control’ than thermal comfort. Air-conditioning was originally designed to control manufacturing conditions so as to ensure the consistency of climate-sensitive products. The idea of mechanically controlling the indoor environment to deliver human comfort arose as a by-product or an unintended consequence of the ability to manipulate the conditions of production. Even today, manufacturing processes frequently takes precedence over the comfort of workers – for example, companies producing wedding dresses in hot countries like Taiwan reportedly provide air-conditioning not for the comfort of their employees but in order to prevent the goods from being stained or marked by the sweat of those who make them.

If air-conditioning is about control, then another relevant consideration is who has the power to decide what temperatures will be and how this influences people’s response when faced with different social and climatic conditions. People who complain about air-conditioned offices, may positively value these same environmental conditions when watching a movie. These and other examples suggest that comfort relates to people's expectations of different environments and to their control over them.

In US homes, air-conditioning was promoted as a method of climate control that would allow the consistent reproduction of the ‘right’ conditions for civilised society. In practice, US manufacturers struggled to sell the concept of an ‘ideal’ climate to the American public. The post-war housing boom was, Marsha argues, more important in accounting for the spread of air-conditioning and the proliferation of cheap air-conditioning units across all sectors of society. Comparison with the UK today is instructive. Since the UK is not on the brink of a massive increase in new housebuilding, it is unlikely that domestic air-conditioning will become established as normal in quite the same way.

Returning to the issue of changing expectations, Marsha explains that in the 1920s when US researchers first defined the ‘comfort zone’ this was set at between 61-65 degrees Fahrenheit. Over 5 decades later the zone has shifted to between 69-80 degrees Fahrenheit, with people now wanting to be warmer in the winter than the summer. In the US today, more than 97 million US households use heating and cooling equipment, but less than 13% of these set their thermostats to what Jimmy Carter famously described as an ‘acceptable’ level in the 1970s. Surveys of UK homes also show steadily rising indoor temperatures –some now rising beyond physiologically determined levels of comfort.

How such trends might be ratcheted back or reversed is difficult to imagine. Marsha suggests that one approach might be to empower consumers and give them control over their environments. Perhaps the real challenge is to change expectations that are increasingly resource intensive to meet. Can American consumers be persuaded to abandon their air-conditioned boxes and get out into the ‘real world’ and if so, what sorts of interventions might this require? Might resistance to air-conditioning actually develop? During recent heat waves, a number of US citizens reportedly experienced ‘cabin fever’ and sought to ‘escape’ from air-conditioned homes and offices. And, while many Americans may now be addicted to coolth, UK households are surprisingly wary of the very idea of an air-conditioned home.

Our next speaker, Tadj Oreszczyn, reflected on expectations and experiences of indoor environmental management in the UK and considered the unintended consequences of well intentioned efforts to improve technical performance and energy efficiency.

Conservatories were originally promoted by building scientists as a passive solar design feature. In theory they provide an un-heated buffer space offering an equable climate at no environmental cost. The results of a study of 5,000 conservatories tell a different story. Nine out of ten were heated in winter, many being used as extra room in which to watch television. More worryingly, some conservatories were air-conditioned in summer - a pattern that is not so surprising given that at least one DIY store gives away a free portable air-conditioner with every conservatory sold. In short, rather than being ‘passive’ buffer spaces conservatories are often actively heated or cooled year-round. With more than 70,000 conservatories built every year in the UK, these trends have significant implications for energy demand.

How did building scientists get it so wrong? The problem, Tadj suggests, is that energy models focus on technical aspects of conservatory performance rather than on the way in which these spaces are actually used. Although there are good technical or economic reasons why people should not like using these spaces all the time - they are for instance generally dreadful in terms of solar glare and acoustics – the evidence is that they do.

This example points to a really fundamental problem for building science. As Tadj explains, in theory, theory and practice are the same, but in practice they are not. In the real world, buildings modify the expectations and actions of those who use them. This is a feedback loop generally missed in modelling exercises that focus, exclusively, on just part of this relationship (i.e. on how buildings are used but not on how buildings influence the way they are used).

Moving on, Tadj noticed that people are continually inventing novel ways of using energy. One such example is in heating outdoor spaces. In one shopping centre in Milton Keynes, over 30 heaters are used to warm an area that is only semi-enclosed. In addition, patio heaters are becoming increasingly common in gardens across the UK. Both cases suggest that expectations of indoor and outdoor environments and boundaries between the two are on the move. Interestingly, the pattern here is not so much one of keeping the 'natural' environment at bay, but of extending the reach and scope of artificial climate control.

Developments of this kind present a real dilemma for UK building regulation and environmental policy makers. What sorts of systems might be devised for regulating energy efficiency given the conservatory's ambivalent status. Are new regulations needed to control or ban technologies (like domestic air-conditioning) widespread use of which will exacerbate patterns of energy demand in future? What research is required to understand and anticipate just how such systems might be used in practice. [2]

The afternoon session offered a change in pace and a chance for other workshop participants to consider how meanings and expectations of comfort have changed and what global warming means for the future of the indoor environment.

The objective of the afternoon session was to collate and compare the views of a range of thermal comfort researchers and practitioners. Each speaker was asked to spend a maximum of 10 minutes reflecting on:

We then asked the audience to classify the positions put forward by each speaker. The results of this exercise are presented in Appendix 1 and discussed below.

As expected, there were significant differences of opinion between those who thought of comfort as a definable condition or as a socio-cultural achievement. As Max Fordham pointed out, human beings need to maintain a core body temperature of 37 degrees centigrade to survive. From this point of view, there is a sense in which human history is a history of more and less resource intensive means of achieving this basic goal. However, there is more to life than simply keeping alive.

Human history is also marked by changing meanings and expectations of comfort. Conditions defined as positively uncomfortable in some cultures and circumstances are entirely normal for others. Most of those who spoke during the afternoon took for granted the view that there were absolute limits to what might conceivably count as comfort but opinions differed on where these limits lay and on whether, when and how conventions of comfort might change.

Workshop participants were also of the view that perceptions of comfort were influenced by factors other than those of temperature, humidity and the relation between indoor and outdoor conditions. Job satisfaction, control over the environment and the extent to which people felt loved or cared for by building managers or co-workers were cited as relevant considerations. Taking these arguments into account, some concluded that the experience of comfort relates to the interplay of many thermal and non-thermal factors. Adrian Leaman was, for instance, able to identify more than 65 such variables believed to contribute to an overall sense of physical and mental well-being.

The afternoon presentations also revealed different ways of approaching the task of specifying and designing comfortable buildings. Design guides and methodologies generally make use of the results of extensive laboratory based studies combined with principles of building physics. Participants were wary of the resulting models, but for different reasons. Max Fordman expressed a general dislike of the way that thermal comfort models obscured the parameters and assumptions around which they were built. In using them, there was a real sense in which designers lost control of what they were doing. Fergus Nicol meanwhile questioned the extent to which such methods constitute a precise science, or whether they offer spurious precision at best. Developing this theme, Fergus argued for a more rugged and robust approach.

In the ‘real world’ experiences and expectations of comfort are shared and expressed in simple terms and in terms that are nothing like as precisely calibrated as those used in the design process. There is, however, evidence that clients are latching on to highly technical specifications of comfort and are consequently expecting unrealistically uniform conditions indoors. It appears there is a gap between client and user concepts of comfort, and those used by designers and manufacturers. This presents further problems, for instance, when trying to figure out what user feedback studies really mean and how, if at all, such information might be put to use in the design process.

The approach taken by Bill Gething was to record summertime office temperatures in order to give an approximation of how people respond when temperatures rise. Although only a relatively crude measure, this provides a basis for thinking about different design strategies – if buildings get very hot for only 10 days a year and if people are at home for half this time anyway, is it ‘worth’ installing an air-conditioning unit? An alternative would be to allow staff to go home on very hot days, expect them to wear shorts, and/or give them a longer lunch hour. In such scenarios, adjusting working practices and social conventions obviates the need for more technology.

More generally, participants expected to see the development of more flexible strategies – whether in relation to the design of air-conditioning, in the definition of standards or in the provision of control. The flexibility of client and user expectations was also important. Max Fordham suggested that designers’ ability to construct naturally ventilated buildings depended on their ability to convince clients that people can count indoor temperatures of, say, 30 degrees centigrade for a few hours of the day as ‘comfortable’ providing they adjust their clothing. Another suggestion was to think about how comfort might be promoted as a lifestyle issue such that non air-conditioned environments become fashionable and desirable.

The extent to which current building regulations support and reinforce certain approaches to building design was considered relevant. So too were less formal but nonetheless influential methods of classifying buildings, for instance describing Grade ‘A’ offices as those with full air conditioning, and rating those with natural ventilation lower down a hierarchy of quality. Many felt that until this situation was redressed, clients would be unlikely to favour naturally ventilated alternatives. Such arguments suggest that the promotion and stabilisation of sustainable buildings will depend on transforming deep-rooted institutional conventions as well as challenging the expectations of individual clients and users.

The air conditioning industry is a significant player in this debate. Andrew Keogh explained that Carrier is considering re-marketing air-conditioning for homes as ‘heating with a bit of cooling thrown in’. By repackaging products, manufacturers hope to overcome the negative associations that air-conditioning has for many UK households. As Andrew also explained, a further task is to demonstrate and develop still more efficient air-conditioning systems.

Buildings generally last a long time. Because of this, many participants felt it was important to consider what opportunities are being constructed in buildings today and what possibilities are being foreclosed. One suggestion was that building scientists might focus on the question of how to provide homes that offer adequate opportunity for adaptation in the future.

As the afternoon session showed, the future of comfort is essentially contested. Many of those present subscribed to a view of comfort as a definable condition, but what were its essential ingredients? On this, opinions differed appreciably. The extent to which variations in meanings of comfort were attributed to cultural expectations, personal beliefs or social norms was also highly variable. Discussions revealed support for the idea that design methodologies should acknowledge physiological, psychological and socio-cultural variation. Others argued that scientific precision was frequently misleading and that different methods of modelling comfort were more and less appropriate for different social as well as climatic conditions.

Views on how and why comfort-related strategies might change in the future were also diverse. Some expected manufacturers to invent and sell new technologies, others thought designers would do more to empower individuals and extend adaptive opportunity. More radical suggestions involved the transformation of cultural convention and institutional habit. Policy options for reconciling future comforts with social, environmental and economic objectives were wide-ranging, again reflecting contrasting disciplinary perspectives and philosophies.

At the end of the first day we had covered much ground and unearthed an impressive diversity of positions and perspectives. During a final round table, we asked participants to comment on what they considered to be the key issues explored in the workshop and to identify ‘sticking points’ for the development of more sustainable indoor environments.

One concern was that certain concepts of comfort are now so naturalised within the language and models of building science that designers rarely question the assumptions on which these are based. Some participants expressed the view that if comfort was redefined to refer to conditions that were socially and culturally ‘satisfactory’ or ‘tolerable’ rather than scientifically ‘optimal’ this would open the way for less demanding design criteria and reduce the risk of ‘failure’.

In relation to the design process itself, a number of participants explained that different actors operated in their own worlds or ‘boxes’ and that lines of communication between researchers and practitioners needed to be opened up. Those involved in building-use studies or feedback surveys complained that designers had a poor appreciation of how people actually used buildings or of the multiple parameters on which their perceptions were based. Meanwhile, designers argued that they required more precise and ‘usable’ results and data that they could really work with.

The difficulty of opening new conversations with clients was seen as a major sticking point. Architects and designers explained that clients base their evaluation of a good building on non-thermal factors, such as cost and prestige. Energy efficiency is low down the list of criteria on which final design decisions are made. A couple of participants made the point that regulatory regimes rarely encourage the construction of environmentally novel or innovative buildings.

Discontinuities between design and management were believed to inhibit good environmental practice. Clients do not always occupy the buildings they commission and often have no control over the way in which systems are used. Furthermore, building service contracts rarely last for more than a few years. This means that the definition and achievement of ‘comfort’ is influenced by multiple interest groups, only some of whom have any incentive to devise sustainable solutions

In focusing on the design of individual buildings, some commented that parts of the bigger picture are missing. Buildings are connected to energy systems and one concern was that utility load management programmes or strategies for infrastructure management are often dissociated from the world of construction. More holistic or integrated thinking might support more imaginative schemes designed to link heating and cooling regimes to patterns of peak or off peak loads. This might produce economic and environmental benefits for building managers, tenants and utility companies.

Buildings form part of the urban landscape and as such it makes sense to make connections between the design and management of the indoor and outdoor environment. The study of comfort has traditionally focused on conditions indoors. A priority is to understand how expectations of comfort vary between indoor, outdoor or transition spaces and how this shapes possibilities for the regulation of thermal conditions in cities overall. Another priority is to understand how conventions and expectations of outdoor comfort are changing. A worrying trend, highlighted by some contributors was that outdoor conditions and weather systems are being judged and evaluated with reference to the increasingly standardised conditions people have come to expect indoors.

Even if building scientists, designers and regulators promote and stabilise more sustainable definitions of comfort this may not be enough to stem unsustainable patterns of consumption. A final point concerns the extent to which the design of indoor environments is driven by changing fashions in the built environment. For example, the latest trend in prestige homes is for massive internal spaces that are particularly demanding to heat and cool.

On the second day of the workshop six academic discussants from different disciplines were invited to reflect upon the ideas and arguments generated on day one, to identify key issues for further debate, to think about the future, and to consider the implications of indoor and outdoor climate change for research and policy.

One was to recognise that definitions of comfort are continually on the move. In predicting the effects of climate change on UK households, it is often assumed that in future people will want similar conditions to those they have today. Results from recent household surveys suggest otherwise. In the UK, people now routinely heat their homes above the accepted standard of 21 degrees centigrade. A key priority is to understand how and why expectations of the indoor environment are changing in this way, and what the direction, pace and limits to such expectations might be.

Research is currently underway to establish how future variations in the outdoor climate might influence conditions indoors. The working assumption is that indoor temperatures will, and should, remain at around 21-22 degrees centigrade. But what if expectations change such that people are comfortable at temperatures ranging from say 15 to 30 degrees centigrade? It is at least sensible to model and explore a range of social as well as technical scenarios for coping with global warming. It would, for instance, be possible to imagine a future in which indoor temperatures fluctuate more widely than they do at present and in which people adopt a much wider range of responsive strategies (e.g. changing clothing).

A second issue concerned the extent to which the ‘goal posts’ were moving within the building sciences. In the past, engineers considered variation in indoor conditions as a bad thing. The challenge was to maintain a uniform indoor environment whatever the weather outside. Now variation is more commonly seen to be good for people and for the environment. It is, perhaps, time to revisit the science of comfort and the assumptions embedded in the programmes of research and analysis on which design methodologies are based.

Engineers and building scientists acknowledge the limits of the models they use, but still pursue levels of precision that cannot be achieved in the real world. Field studies conducted since the 1940s have shown that human ‘expectation’ influences perceptions of comfort. Since it is difficult to explain or quantify this ‘factor, it has been set to one side and marginalized. The development of ‘super PMV’ – a revised and extended version of Fanger’s original thermal comfort index – aims to incorporate and account for differences in expectation. However, the prospect of folding a new variable – “culture” – into a methodology that is itself the product of a particular social and cultural context is clearly problematic.

Manufacturers’ efforts to promote the uptake of air-conditioning in UK homes contribute to the continual renegotiation of comfort. The types of air conditioning units being specified and sold will help define what comfort means to households in the future. Patterns and processes of diffusion are also important in determining the extent to which definitions of indoor comfort change. If UK house builders and property developers incorporate air-conditioning as a ‘value-added’ feature of UK homes, will this ensure its rapid diffusion across all parts of society? How, if at all, might environmental regulators control the diffusion of centralised or stand-alone air-conditioning systems?

If we view the making of indoor environments as a systemic process – that is, as one involving manufacturers, designers, urban planners, occupants, governments and regulators – it is immediately apparent that opportunities for heating and cooling are multiply configured. Air-conditioning manufacturers and designers of environmental controls, including thermostats, set the limits within which systems can be operated. Utility companies provide the network capacity required to run heating and cooling systems (again up to at a certain level). Designers and house builders specify and construct indoor spaces of various sizes and equip them with different adaptive opportunities. Consumers fill these spaces with differing degrees of coolth and warmth. Environmental policy making needs to recognise that indoor environments are constructed by a variety of supply chain actors. The future of comfort (and of associated patterns of energy consumption) is not simply determined by human physiology, by user ‘behaviour’ or by the ambitions of clients and designers. It is, instead, better understood as a consequence of their mutual interaction.

The two-day workshop helped to clarify four contrasting, but not mutually exclusive, possibilities for the future of comfort.

Rather than expecting a very narrow range of indoor environmental conditions, all year round, people may become used to greater variety such that they expect to be colder than at present during the winter and warmer than at present during the summer. Seasonal fashions would really come into their own and would provide an important means of managing climatic variation and accommodating anticipated forms of global warming without adding to the problem itself. More ‘elastic’ definitions of comfort would significantly reduce resource consumption. Designers would still be expected to produce buildings that can be heated and cooled but only by as much as required to keep indoor temperatures within an expanded range.

The variety of actual indoor environmental conditions is already impressive. Although most commentators expect meanings of comfort to converge around the globe (partly as a result of designers’ reliance on standardised methodologies, partly because of the powerful commercial interests at stake), this is not the only option. It is, for instance, possible to imagine the reinvention and positive valuation of local cultural and climatic variation. Again this would reduce the environmental cost of comfort.

A third possibility is that conventions of comfort and clothing stabilise but greater effort goes into finding much more efficient ways of providing and delivering precisely defined conditions of ‘comfort’. This might involve new forms of technology, better controls, or carefully calibrated, climatically sensitive, passive design strategies.

Finally, it is at least possible that interpretations of comfort will develop in ways that are even more demanding than those of today. What if people - for one reason or another - expect to be even warmer during the winter and even cooler during the summer?

All four routes are possible and all are being pursued by different actors. The first finds favour with more radical environmentalists and with those who believe that buildings should be designed to maximise adaptive opportunity. The second is gaining ground theoretically, but has a long way to go given the dominance of the physiological paradigm. The third is where policy lies now, and is also a model promoted by the air-conditioning industry. Though not explicitly promoted as such, there is some evidence to suggest that the fourth possibility is becoming a reality.

All of these trajectories are likely to co-exist in future. What will happen overall, will be an emergent outcome of the intersection between them and will depend on which pathways dominate, and on patterns of convergence or divergence over time. From an environmental point of view, the challenge is to recognise that each affords different opportunities for policy intervention.

To conclude, we make some suggestions about the sorts of directions that research and policy might take in order to meet the challenge of constructing more sustainable forms of comfort.

The development of more efficient methods of heating or cooling will not guarantee a reduction in energy demand. The challenge of establishing more resource efficient concepts of comfort requires a much more sophisticated understanding of how people actually use buildings and how technologies shape what people do. Studies of buildings in use may help in developing policies that reduce energy consumption in fact as well as in theory.

Thermal comfort research has tended to focus on the office environment. In practice the home is likely to become an increasingly important site of global indoor environmental change. New questions arise in this context, not least because issues of ‘productivity’ and ‘satisfaction’ take on new meaning. Again most work has concentrated on comfort during the day and when people are up and about. But what of the night time? What does it mean to have a comfortable night’s sleep? Marking out the bedroom as an area for special attention might help to anticipate changes in the demand for night-cooling and highlight the range of strategies that might be enrolled (e.g. opening windows, using fans).

The chances of developing and institutionalising more elastic and more forgiving concepts depend, in large measure, on drawing together and presenting cultural and historical evidence of the range of conditions in which people are comfortable. Ethnographic studies are therefore of global as well as local significance, especially those that show how different concepts of comfort have fallen in and out of favour, and those that track the social and cultural dynamics involved.

Fashion, fabrics and non-building related strategies have a role to play in the provision of future comforts and should be taken seriously. The cultural and historical relation between clothing and environmental control requires further investigation.

In addition, it is important to understand the development and diffusion of air-conditioned ways of life. In comparing the spread of air-conditioning in the US and the UK, the significance of cultural values and contrasting forms of ‘resistance’ becomes apparent but only in the most general of terms. More needs to be known about the detailed dynamics of indoor environmental change. Those who want to stem the spread of air-conditioning - especially to homes in the UK - might do well to think about ways of mobilising consumer resistance, for example tapping into long-standing ideas about health, well-being and fresh air.

Most importantly policy makers should realise that the future of comfort is not entirely fixed. It remains a matter of contention. By actively promoting debate about the indoor environment and associated ways of life it may yet be possible to avoid locking-in to social and technical trajectories that really are unsustainable.

The table records participants’ views of how each afternoon speakers (listed in column 1) responded to our three questions concerning: the conceptualisation and definition of comfort (column 2); the factors explaining change in what comfort is (column 3); and, the implications of changing definitions of comfort for environmental policy and climate change (column 4).

[1] Our typology of comfort paradigms and a description of each position can be found in the accompanying paper on “Comfort paradigms and practices” (see Chappells and Shove, January 2004).

[2] These and other questions are being addressed as part of a project Tadj is co-ordinating on ‘Domestic Air-conditioning’.