On the Need not to be Constructivist about Affectivities


This is the slightly revised text of a paper I held at the Parsons School of Design at the New School, New York, in January, 2009


In this paper, I pursue both a first-order and a higher-order goal. The first-order goal is to point out what seems to me to be a weakness in Elizabeth Shove’s account of such everyday behavioural practices as the thermal management of working and domestic environments, the laundering of clothes and the bathing of human bodies. The higher order goal is to identify, through this critique, a crucial task whose accomplishment is required if we are to profit from Shove’s central practical insight. The insight is that achieving a more sustainable order primarily requires what she calls the reconfiguration of everyday routine behaviour. The philosophical task required for it to bear its political fruit is the philosophical explication of what it is to be the kind of moderately rational, self-conscious beings we are.

Shove’s work on the ordinary, everyday use of ordinary, everyday things provides a healthy corrective to certain popular conceptions of the causes of unsustainability. Shove points that

  1. problems of sustainability are not simply failures of moral or prudential will to be addressed by getting people individually to engage in different kinds of behaviour, e.g., driving less, saving water, using eco-friendly light bulbs, etc. – pace much light green environmental politics of the kind popular in current government and business circles;
  2. the causes of unsustainability do not lie solely in the acquisition of things for purposes of shoring up identity or advertising it to others, pace existentially or semiotically psychologising accounts of the consumer society; and that
  3. the causes of unsustainability do not lie solely on the production side, as if they could be addressed solely by smarter technology and better regulation, pace many in government and business circles, and indeed those who claim to be promoting radically new concepts of design, e.g., holistic systems theorists, natural capitalists, bio-mimeticists and the like.

No doubt identity- and status-driven consumption plays its part in the overall unsustainability of our current existence. But the main causes of unsustainability lie in habituated everyday behaviours about which no one can get existentially or semiotically aroused. They lie in the way we regularly eat and cook, keep ourselves and our things clean, move about our world and keep ourselves warm or cool. One cannot seriously or permanently change these unsustainable routine behaviours simply by moral or prudential critique since merely exhorting individuals not to engage in old behaviours is not to give them new ones. In the absence of concrete alternative ways of living well exhortation can only bring about an unstable condition of exception always threatening to lapse back into unsustainable normality.

Nor indeed can one trick these routine behaviours into assuming more sustainable form simply by re-designing the technological means they use or the regulatory framework within which they operate. Behavioural practices involve understandings of what one needs to do and in particular what one uses in order to do them. And these understandings of need and how to satisfy it help to determine the concrete shape which ways of using a technology assume within a regulatory framework. So these understandings can in fact work to undermine the potential for sustainability inherent in technologies and regulatory norms. As Shove puts it, these understandings constitute a conception of the service offered by technologies. And “(i)t is,” she says,

impossible to think about transitions to sustainability without also considering the redefinition of service. In this case, as in others, it is the relation between resource efficiency and the simultaneous respecification of demand that counts in practice. (Shove 2003, p.138)

So as well as designing more sustainable technological means and regulatory frameworks, we must also consider users’ expectations as to what technologies legitimately yield and regulations legitimately constrain. The expectations driving a behavioural practice, no less than the technologies used in it and the regulations governing it, need to be, as Shove puts it, respecified or reconfigured.

Yet as important as this point is, there is, I think, a problem in Shove’s account. We can bring this problem out by asking what it is more precisely to reconfigure the expectations, beliefs and desires implicated in such everyday practices as bathing and laundering. When Shove comes to address this kind of question, the results are thin. At one point,1 she suggests that in order to shift everyday routines of movement and transportation away from the private car, one might reduce the supply of roads and highways. These, after all, have made mass driving possible, hence created the car culture. Similarly, in order to shift everyday routines for managing the thermal environment, one might re-configure the ideas and images which have both encouraged and themselves been entrenched by current practices of heating and cooling buildings. Thus, various governments have attempted to encourage less use of air-conditioning by promoting a smart business look without a tie and standard business suit.2

The refusal to supply roads, as bold as it is, is just a negative measure which does not on its own constitute a general strategy of weening people off the use of cars – precisely of reconfiguring current practices of moving about. So taken on its own, it does not acknowledge Shove’s requirement that one seek to reconfigure underlying expectations of service, in this case, of mobility. The suggestion that one revise notions of business smartness to enable more sustainable thermal management does acknowledge Shove’s requirement. But the reconfiguration it embodies is a mere tinkering at the edges of current conceptions and expectations of normal business behaviour. So this reconfiguration does not fundamentally alter the demands currently placed upon thermal management. It falls within the pale of standard ‘environmental education’, hence surely could not generate the kinds of radical reduction in energy intensity needed for significant gains in sustainability. Neither suggestion distinguishes itself, therefore, as a practical proposal to which one is led only because one endorses Shove’s general theoretical position on the nature of everyday behavioural routines and their status as the most fundamental sources of unsustainability. Yet surely Shove’s position must yield proposals for public policy and political action which are genuinely distinctive of this position.

I. Towards Identifying the Deficit in Shove’s Account

Why, then, does Shove’s theoretical position appear only able to generate thin guides to action which are not particularly distinctive of it? This poverty seems to me to be due to a deficit in the conceptual framework which Shove at least appears to employ when describing the synchronic process and diachronic evolution of behavioural practice. In order to identify this deficit, let consider a number of key passages from Shove’s book.

Shove writes,

Taking practices to be themselves constitutive of the ends they purport to serve, Hackett (1993) argues that rather than being a predefined goal or something that people strive to achieve, cleanliness is best understood as the outcome of whatever it is that people do in its name. Following this kind of reasoning, it is the everyday activity of laundering or showering that convinces people there is dirt to remove. (Shove 2003, p.85)

In this passage Shove is clearly endorsing Hackett’s views according to whom practices define the ends they serve. Indeed, so much is this so that practices might be described as self-fulfilling. If, after all, the end of a practice is defined by it, then it cannot stand over and against the practice which ostensibly serves them as that against which the effectiveness of the practice could be evaluated.

Now taken literally, this kind of claim is counter-intuitive. Surely cleanliness and dirt are independent features of the world which different practices can engender or eradicate with different degrees of success? Shove concedes and in some measure seems prepared to accommodate this point when, several pages later, she says,

In appreciating that meanings of cleanliness and dirt arise and are given expression through things and practices I also recognize that things and practices have ordering and dirt-defining properties of their own. (Shove 2003, p.90)

But Shove never makes clear how one can consistently maintain both that there are independent notions of cleanliness and dirt and that the practices and means of achieving cleanliness and removing dirt determine the meaning of the terms ‘clean’ and ‘dirty’. And she soon allows this concession that there are independent notions of cleanliness and dirt to slip out of the picture. For now Shove says

In bathing and showering as they do, people manage [the concerns driving their behaviour] in their own terms, tapping into a contemporary repertoire of reasons and rationales in support of their actions. It is, however, important to qualify this sense of ‘external’ pressure. After all, routines and practices are themselves constructive and constitutive of collective convention. Normality is made of what people normally do – hence Jane, Tim, Angela, David and Sarah convince themselves of dirt and cleanliness, of being relaxed or awake, of the rigours of the outside world and of being ready to face them, through the way they bathe. (Shove 2003, p.115)

Notice how the concession made in the passage previously cited is here almost entirely taken back: people bath and shower as they do for all sorts of different reasons, but these reasons are a contemporary repertoire which bathers and showerers have acquired, not externally, but, or so it seems, through the practice itself. Shove adduces the testimony of the interviewees Jane, Tim, Angela, David and Sarah in order to corroborate this point. These individuals all learn what it means to be dirty or clean, relaxed or awake, and being ready to face the rigours of the day, through their everyday practices of bathing. And from this Shove concludes that we should not think of these individuals as possessing independent notions of cleanliness and dirt which justify their perception of how they bathe and shower as effective ways of removing dirt, getting clean, relaxing, waking oneself up, or preparing oneself for the coming day. Not only do they come to understand cleanliness and dirt in the way they do through actually engaging in the practice; this practice constitutes what they understand. Jane, Tim, Angela, David and Sarah do not bring any desire to remove dirt, get clean, find relaxation, be wakened up or get ready for the day into everyday practices of bathing.3 Rather, they acquire these desires and motivations only in the practice. And this is because, claims Shove, it is only in the practice that the relevant concepts get defined.4 These concepts are, she says, emergent properties of the practice – see Shove 2003, p.138.

Given all this, it comes as no surprise to find Shove making the following very strong claim:

When asked about what is involved in producing clean clothing, virtually all those interviewed in the Unilever study agreed: anything that had been through the washing machine was, by definition, clean. Items emerging still stained or marked might be discoloured or disfigured but they were not dirty. When asked ‘what has to be done in order for it to be clean … what actually has to happen, for you, for it to be clean?’ respondents had no hesitation. This reply is typical: ‘For it to be fed through that washing machine, that is it.’ (Shove 2003, p.147)

Imagine you are asked what has to be done in order to fix a flat tyre, what actually has to happen, for you, for the flat tyre to be fixed. You would say that the car has to be jacked up and the flat tyre replaced by the spare. You might, of course, say something different, say, that the car has to be jacked up, the tyre repaired and then bolted back onto the brake hub, or perhaps replaced entirely by a new tyre. And the very fact that you could say these different things shows there to be one thing you would not say: that for you, for a flat tyre to be fixed was, by definition, for it to have been replaced by the spare. You do not at all understand the question asked of you as one concerning your concept of what a flat tyre is and when it is fixed. Rather, you understand as a question about what to do in order to get a flat tyre fixed – naturally, of course, what one would typically do to fix it, but not, of course, necessarily, since as we have seen you acknowledge the possibility of fixing the flat tyre simply by buying a new one.

Yet Shove takes the strictly analogous question put to interviewees in the Unilever study to be a question about the very concepts of cleanliness and dirt – in which case the answers given to her question would commit their givers to the false view that necessarily, in all possible worlds, to get clothes clean5 one had to put things through a washing machine. Shove is forcing a false interpretation upon her interviewees’ answers and she appears to be doing so in order to confirm a theoretical conception she has convinced herself of in advance. According to this conception, practices define the ends they serve; they construct the concepts, hence expectations, norms and standards which practitioners wield when participating in them.

To point this out does not, of course, show that the conception Shove seeks to confirm is indeed false. It merely shows that the responses Shove has got in her interviews do not provide the confirmation she thinks they do. One might therefore still have good theoretical reasons for endorsing Shove’s view of the ends served by a practice as internal to it. Furthermore, one might also feel that there are good practical reasons for embracing Shove’s constructivist account of the ends of practice. Surely, one might argue, if the goals and interests we pursue when bathing, laundering and, mutatis mutandis, managing the thermal properties of our living environments are internal to these practices themselves, then they are as changeable, hence malleable as these practices themselves. To recognise the socially and historically constructed nature of these goals and interests is therefore to recognise that they are not set in stone. This permits one to countenance the possibility of more sustainable alternatives to current practices.

In what follows I will argue that there are no good theoretical reasons for endorsing the constructivist conception of the ends of practice implicit in Shove’s account. In addition, I will argue that this is a desirable result. For this constructivist view does not at all facilitate, it limits the potential of Shove’s account to provide distinctive, practical guidance.

II. From Ends to Desires to Affectivities

Let us grant, for the sake of the argument, that modern day launderers literally mean by clean having gone through the washing machine. But what is it to go through the washing machine? Is it to take one’s clothes and let them sit overnight in the washing tub? Obviously not. One has to turn the power on, turn the knob to select a suitable programme, put suitable quantities of washing powder and bleach into the machine, set the whole lot going and wait until it turns itself off. If one fails to do any one of these things, and if the machine refuses to cooperate in any of these steps, then one has not fed the clothes through the machine properly , i.e., in accordance with its function, which is to get clothes clean. With this, we see that it makes no sense to maintain that by ‘clean’ contemporary launderers or anyone else could mean ‘having been fed through the washing machine’. The latter notion in fact presupposes an understanding of what it is to be clean. It could never, not even in principle, occur as the definiens of the definiendum ‘clean’.

But this is mere dialectic. A more powerful, because more positive argument can be derived from some cases adduced by Shove herself. Consider Shove’s interviewee, Jane, an American university student who

(a)s a teenager, … had lots of long, thick hair and a set of expectations that obliged her to shower for thirty minutes during an hour-and-a-half ritual of ‘getting ready’. Appearance was a central concern and the nature of her unruly hair was such that it ‘needed’ shampoo, conditioner, detangler and more. Unnoticed until early adolescence, Jane’s skin also demanded special treatment. Jane’s elaborate yet necessary morning routines were thrown into disarray by a brief but significant school exchange to water-short Spain. Fearing the excesses of their American visitor, her hosts provided clear instructions on how to kneel in the bath and make the most of just one bowl of water. To Jane’s surprise she got used to this new regime. Although still committed to detangler and conditioner, she tells this story by way of explanation. She now showers just three times a week, never for longer than a few minutes, and she really can’t remember when she last had a bath. (Shove 2003, pp.109-110)

This seems to illustrate the kind of point Shove wants to make about needs, namely, their socially constructed character: Jane initially has all sorts of resource-intensive routines for pandering to her teenage vanity, needs imposed upon her by society and which, as Shove’s use of scare quotes indicates, are not really needs at all. She goes to Spain, where she finds herself forced to participate in a considerably leaner bathing regime. Eventually, she acclimatises to this new regime and as result finds, to her own surprise, that many of her old ‘needs’ have gone.

Note now how Shove describes this case – as if Jane’s transition from a water- and energy-intensive regime of bathing to a less intensive one were like the transition from driving on the left hand side to driving on the right hand side of the road. There is surely something wrong here. Jane’s transition undoubtedly involved an element of re-acclimatisation and re-habituation. But much more was also very likely going on here and this is obscured by Shove’s tendency to assimilate this transition to that an Australian driver has to undergo when moving to live and drive in Germany. For when Jane gets back to America, she does not revert back to her old water ways. She is precisely not like the Australian driver who, having spent several years in Germany, reverts back, perhaps with some initially difficulty, to driving on the left hand side. For she has learnt something about her former ways – not just that she can live without them, nor even that she ought to live without them (because perhaps she knew this latter before she went), but rather she has learnt how to bathe better in a sense of the term ‘better’ which connotes a more satisfactory synthesis between looking and feeling good and doing the right thing.

In other words, Jane has reconfigured an aspect of how she is affected by the world, namely, her initial adolescent sensitivity to the opinions of others concerning her appearance. Crucially, this is not at all to say that she has simply lost her initial concern for, and sensitivity about, the opinions of others. As Shove herself reveals, Jane remains “committed to detangler and conditioner.” She is still, therefore, concerned about her appearance but now her concern is tempered, or so we may plausibly assume, by recognition of the ethical implications of intensive water use.6 What permitted Jane to learn how to temper her behaviour by this recognition was the experience in Spain, an experience which, we may assume, taught her, not so much that she could live well enough without flouting ethical requirements, but rather how to do so. More precisely, Jane learnt that she could live well enough without flouting ethical requirements by learning how to do so. So Jane has not merely a reconfigured, she has an improved way of being affected by the opinions of others concerning her appearance – she can, as one says, now put these judgements into their proper perspective. And with this ethically improved way of being affected comes an improved understanding of how she had formerly responded to the judgements of others: whereas she had perhaps previously thought of her former bathing behaviour as harmless and normal, hence permissible, she now sees it for what it is, namely, a combination of self-indulgence and insecurity.

Note now a crucial implication of this clearly possible, even plausible account of Jane’s transition from water-intensive to less water intensive bathing practices: Jane has an improved concept of what it is to look good or even ‘cool’, one now defined not in terms of what others will actually say but rather in terms of what they should say. Jane is happy enough to make concessions to the opinions of certain factual others and even to her own vanity, but now not slavishly so, at the cost of relevant ethical considerations. If, however, this is so, then the concept of looking good, even ‘cool’, i.e., adolescently fashionable, can never be defined in terms of any conventional process or practice Jane might put herself through, whether the old process of a thirty minute shower or the new process of using a bowl of water. In order accurately to describe Jane’s transition from less to more sustainable practices of bathing, we must precisely not say that for Jane to acquire an elegant, socially acceptable, even ‘cool’ appearance just is to use certain specific items of technology in certain rituals of bathing and making oneself up. Any such view of Jane’s transition prevents us from seeing it for what it is and indeed what Jane herself might take it to be, namely, a process of moral growth and maturation.

Already this suggests that if one takes the kind of constructivist view Shove at least seems to recommend, then in at least two respects we find ourselves in serious conceptual trouble. Firstly, at the level of theory we find ourselves unable to capture certain kinds of transition from one behavioural regime to another – those kinds, namely, in which the individual or individuals involved improve morally or at least mature in some respect. The conceptual framework Shove offers is simply unable to capture this kind of transition because in order to capture it, one must not construe the ends of a practice – in Jane’s case, personal appearance – as internal to the practice. In order to capture this kind of transition, these ends need to be reconfigurable in the sense of being able to preserve and assert themselves in better, more context-sensitive adaptations to requirements external to the practice – in Jane’s case, presumably ethical requirements concerning the use of water. Since this new synthesis, in which old ends are both preserved and transformed, lies beyond the original practice, so, too must the ends themselves.

Secondly, if we take a constructivist view of the ends of practice, then, because we cannot conceptualise Jane’s kind of transition theoretically, we cannot develop policies and strategies built around encouraging it. Notwithstanding her evident intentions to the contrary, Shove’s conceptual framework does not permit one to acknowledge something in one way obvious, yet still to be learnt by many concerned about environmental issues. Typically, programmes of environmental education only improve levels of environmental knowledge, not levels of environmental behaviour. Part of the reason for this is already implicit in Shove’s work, namely, that it is not enough to fill even the most well-meaning of individuals up with knowledge about the disastrous prudential and ethical implications of their behaviour. Individuals must also be provided with the opportunity to work out, more or less for themselves, ways of accommodating their desires with the requirements of prudence and ethics. And, as the case of Jane illustrates, these desires can change while remaining the same, as the ends desired evolve, both accommodating themselves to the demands of prudence and ethics yet continuing to claim their due. Jane’s story, at least as elaborated here, illustrates this process of coming to a new accommodation of ends and desires with ethics and values – a new accommodation in which one and the same end, hence one and the same desire for this end, assume a new and more mature form. But if we take seriously the constructivist claims Shove makes, we find ourselves unable to see this. Unsurprisingly, then, the practical conclusions Shove draws from her analysis are thin.

These two problems are, I think, the relatively superficial manifestation of a deep fault in the way Shove thinks about purposive behaviour and behavioural practice generally. Jane’s water-intensive bathing, we may assume, was a prophylactic response to potential criticism and rejection by others; she desired the end of an elegant or ‘cool’ appearance because she was affected by the possibility of negative judgement and rejection by others. Such judgement and rejection really hurt her. This indicates a crucial point: one can only have desires, interests and ends, hence engage in purposive behaviour, insofar as one is affected in various ways by the world. The English philosopher G. E. M. Anscombe once demonstrated this point rather well when she asked us to consider how we would respond to someone who expressed, in all apparent sincerity, the desire for a saucer of mud.7 We would, she observed, respond to this expression of desire by searching around for some way of making possession of the desire expressed and its goal intelligible to us. Thus, we would speculate whether the person wanted a saucer of river mud because this person wanted to enjoy its rich, dank smell – much as one might enjoy the smell of the Australian bush, heavy as it typically is with the scent of eucalypts. Note what we are doing here: we are attempting to find some way of being affected by the world which we find intelligible because we ourselves can be affected in a sufficiently similar way. Once we have identified some such affective disposition,8 the strange desire and the strange purposive behaviour in which it issues all fall into place as intelligible. In general, purposive behaviour, desires, interests and ends all presuppose that what has or engages in them is disposed to be affected by the world in appropriate ways.9

Now Shove’s constructivist commitments insinuate a particular ontological conception of affective disposition or, as I shall sometimes also call it, affectivity. Consider Angela, another of Shove’s respondents. Angela is made to squirm by the thought of showering

any less than twice a day, more in case of exercise and physical exertion … . Sweat and its removal is her main concern and it is this that determines and gives meaning to when and how she washes. She will not exercise without the prospect of a shower ahead. Likewise, only a shower will do for the purpose is to get clean and this is not, in Angela’s view, something that can be done in a bath. The idea of sitting in a pool of dirty greasy water is essentially revolting and the notion of washing her hair in this liquid is quite beyond the pale. (Shove 2003, p.111)

Angela is disposed to experience feelings of disgust in the presence of sweat, in the first instance her own but conceivably also the sweat of others. This disposition to be revolted by the presence of sweat underpins and makes sense of her various desires and the behaviours in which these desires issue.

Shove’s comments on efforts to define scientifically levels of thermal comfort10 suggest that she endorses the following general picture of Angela’s affective disposition: a way of being affected by the world, whether feeling disgust at the presence of sweat or revulsion at the thought of drinking recycled water,11 is nothing more than a disposition to be caused to have certain powerful feelings or emotions by certain objective facts. Crucially, the causal connection between the subjective feelings and the objective facts is seen as utterly contingent. That is, on this picture, the affective response which an individual displays in the face of the objective facts which cause it cannot be grounded, or made sense of, in terms of what these facts are. This reflects, of course, the constructivist conviction that the disposition to give such and such an affective response to such and such objective facts is simply the result of socialisation into the relevant practice.

But is it the right way to think about affectivity? In particular, can it accommodate the point made above that purposive behaviour, desires, interests and ends all presuppose affectivities because affectivities make it intelligible, hence possible for an agent to have desires, interests and ends, hence to engage in purposive behaviour? In order to preserve this point, might we not have to understand the subjective feelings as not just caused by certain objective facts but also as rendered intelligible by them? In which case these objective facts could not be simply natural scientific facts but rather certain pre-scientific, perceptible facts of everyday life?

III. The Eliminativism in Constructivism and its Incoherence

It is not hard to see that a constructivist interpretation of such behavioural practices as bathing and laundering is committed to a version of what is known in philosophy as eliminativism. To be eliminativist about a certain kind or range of phenomena X is to insist that, whatever people might think pre-philosophically, X’s do not in fact exist; the concepts people wield of being an X are in fact empty, hence will be eliminated from our theoretical account of the world as this progresses towards the truth. In its original and classic formulation eliminativism concerns mental states and events in general: the eliminativist argues that talk about beliefs, desires, perceptual experiences and the like is very much like talk about witches and fairies. Even though we currently talk of beliefs, desires, perceptual experiences and other such mental states and events, there really are no such things and so talk of them will eventually disappear, at least from our theoretical account of the mind, if not necessarily from ordinary everyday discourse, where such talk might prove to be practically indispensable.

As we have seen, the constructivist interpretation of practice claims, at least implicitly, that affective response is just a matter of responding in certain conventionally instilled ways to certain objective facts which, in and of themselves, do not legitimate any one way of responding affectively to them than any other. For according to the constructivist there is nothing beyond the practice in which the relevant range of affective responses arises that could show this practice, and thus this range of affective response, to be the ‘right’ one. So the only things beyond the practice are those natural scientific ascertainable facts which cause the diverse affective responses characteristic of the diverse practices.

Take, for example, body odour. In no sense are there, beyond the ordinary, everyday practice of avoiding excessive levels of body odour, any objective facts which license the evaluations participants make in the name of this practice. The only objective facts lying beyond practices of bathing are whatever natural scientific facts cause the olfactory responses by which this practice maintains itself – patterns of olfactory response which the practice has itself engendered. These natural scientific facts are presumably such things as a mixture of certain chemicals exuded by the body, or perhaps by bacteria on the body, under certain conditions.

By contrast, everyday characterisations and evaluations of people as smelling abominably, having terrible body odour and the like do not pick out anything genuinely existing in the world. For there are no such things as abominable smells and terrible body odour; like concepts of witches and warlocks, these concepts, too, apply to nothing, however much those who wield them think they do apply to something. The constructivist must say this for the following reason: if these characterisations really did apply to something, then the historical and cultural diversity of bathing practices would force one to distinguish between practices which did, and those which did not, wield these concepts correctly. But just this is, according to the constructivist, impossible. The constructivist is committed to an eliminativism about affective response, or rather, the objects thereof.12

Does it suffice, however, as a characterisation of the ontological constitution of affective response to say that when we find ourselves confronted by someone with extreme body odour, a certain array of chemicals exuded by the person causes within us a certain distinctive subjective feeling, say, of revulsion and disgust? I suggest not. Such subjective feelings or affective responses are defined in their very identity by their character as responses to or towards something, viz., body odour. One must understand them as such because only so can one grasp their functional character, that is, their character as causal dispositions which incline their possessor to various forms of intentional behaviour, e.g., moving away from the offending individual, telling this latter to have a wash, etc.13 The behaviour caused by affective response is, of course, purposive; it is therefore subject to conditions and criteria of rationality. But in order to be thus subject to these conditions and criteria, hence to fulfil them, the behaviour caused by an affective response to the world must also be intelligible in the light of this response. Relatedly, such behaviour can only appear to both actors themselves and interpreters of them as thus subject and as fulfilling the conditions and criteria of rationality if it appears intelligible to them in the light of the affective response which causes it. A person’s angrily telling another to have a wash is only intelligible as purposive behaviour subject to conditions and criteria of rational assessment because it is intelligible in the light of the revulsion and disgust which caused it.

But if this is so, then one must be able to grasp what causes the subjective feelings and affective as similarly intelligible. Angrily telling another to have a wash is only intelligible in the light of the revulsion and disgust which causes it because the revulsion and disgust are themselves intelligible in the light of what caused them, namely, the other’s offensive body odour. An affective response must indeed constitute an intelligible, understandable response towards something, in this case, body odour. Only for this reason is it able to render intelligible the purposive behaviour they themselves cause. No description, then, of what it is to respond affectively to something, in this case, body odour, can be satisfactory unless it builds into its account recognition of the fact that affective response is rendered intelligible or understandable by what causes it. In short, in order to be able to see the purposive behaviour caused by affective response for what it is, namely, intelligible in the light of this affective response, one must also see the response itself as intelligible in the light of whatever causes it. But then what causes it must from the outset be understood as falling under descriptions or characterisations in the light of which the affective response caused by it makes sense.

Now body odour is, I am presuming, a certain mix of chemicals exuded by the body or bacteria on the body.14 But whatever natural scientific descriptions of it happen to be true, none of them render the affective response intelligible. To know that a person’s nose is being currently bombarded with a certain array of chemicals is not in the least to know why it makes sense for the person whose nose it is to feel revolted and disgusted. The ontological constitution of affective response is therefore such that from the outset it is not and could not be simply a response to certain objective facts. More accurately, it is not simply a response to objective facts when by an objective fact one means some bit of reality characterised in purely natural scientific terms, or at least in terms available to beings who did not have distinctively human capacities for being affected by the world, e.g., someone from Alpha Centauri who had no olfactory capacities. If this is how affective response were, it would be impossible to ascribe, whether to another or indeed to oneself, either the response itself or the intentional behaviour it causes. In other words, it would be impossible to recognise either another’s or one’s own affective response and ensuing purposive behaviour – something manifestly false, indeed absurd. To be eliminativist about the objects of affective response is therefore literally incoherent.15 And this incoherence must be inherited by a constructivist account of practice since this account is, as we have seen, tacitly committed to such eliminativism.

IV. Towards a Fuller Picture of Rationality and Agency – The Philosophical Task

I have already argued that a constructivist picture of affective response prevents us from conceptualising precisely the kind of transition of interest for the politics of sustainability. This is the kind of transition we are presuming Jane to have made while in Spain. The discussion of the previous section has, however, put us in a position to make a stronger claim: if one is constructivist in one’s conception of affective response and its objects, one will simply fail to see the positive role it plays in the kind of transition Jane made and indeed in rationality itself. In fact, the reason why Shove has trouble conceptualising the kind of transition Jane makes, hence can say little about it, lies in the way her implicit eliminativism prevents her from seeing affectivity as not just an obstacle to be overcome, but precisely the very stuff and target of transitions to better behavioural routines. To this extent, the eliminativism inherent in her constructivism prevents her from seeing how affectivity is implicated in the very structure of rational deliberation and action.

As we saw above, Jane’s concern for elegant, even fashionable appearance survived the transition from profligate American to leaner Spanish bathing habits. Let us investigate a little further this capacity of one and the same end of behavioural practice to exist, with appropriate modifications, across different behavioural practices. What does this tell us about the affective responsiveness in which desire for, or interest in, an end or goal is grounded? As a self-conscious subject, one obviously bears a certain ethical responsibility for the acts to which one is motivated by how the world affects one. Body odour can be nauseating, revolting and evocative of at least momentary contempt and loathing. But if the offending person has not washed for three months because he is a refugee fleeing oppression, then one must not allow such feelings to shape behaviour in the fashion they render intelligible.

Now in what way might a constructivist conception of affective response truncate our understanding of how individuals can exercise the control over their affective responses sometimes ethically required of them? The character of affective response as subject to ethical constraint implies that, to one degree or another, the affects evoked by the world in distinctively self-conscious selves are under the context-sensitive, at least moderately rational control of these selves. Confronted by someone who, having been on the run for months, is seriously unwashed, one does not give free rein to one’s affective response and dash out of the room. One stays there to help him, the unpleasantness of the situation notwithstanding. One just grits one’s teeth and tries to master the feelings of nausea and even indignation which extreme body odour can call forth.

But is this exercise of ethical discipline over one’s affective response just a matter of gritting one’s teeth and bearing it until one becomes so habituated that one’s very capacity to notice the odour disappears? Is one’s affective responsiveness nothing more than a wild beast, struggling to assert itself over one’s morally informed will, which must therefore struggle against it? In which case relief from this situation of conflict can come only through killing the wild beast off, that is, through losing one’s responsiveness? Surely not. Firstly, it is not true that after a sufficiently long period of clamping down ethically on how one responds to the refugee, one’s initial affectivity just goes away. Repeated experience of something may certainly desensitise one to it but it is wrong to describe this as a matter of losing one’s initial sensibility. In the kind of case under consideration, one never loses the capacity to perceive what one perceives as revolting, at least in the sense that one remains able to perceive it as having that sensuous character which had once so revolted one that one was inclined to flight. What gets lost is not the capacity to notice these kinds of thing, but rather the capacity of the capacity to notice to break uncontrollably into the course of action and send it off in the wrong direction. One surely always remains able to notice the sensuous character of the entities perceived and this character may even in a sense continue to revolt one.16 One has, however, acquired the capacity to prevent it revolting one in the sense of rendering one unable to act as the situation requires. This is confirmed by the fact that in situations of extreme stress or weariness one may break down and the revulsion comes flooding back. True, when this capacity to restrain has become second nature, one is aware of the sensuous character of what one perceives merely as something one had initially found revolting. But even in this case, awareness of it, this same sensuous character, is still there. This would not be possible had one ceased to notice in the sense of literally losing the capacity to notice.

Secondly and more importantly, the insinuation that affective response is a wild beast that the will can only struggle against and hopefully kill off treats affective responsiveness as at best incidental, at worst inimical, to rationality and self-conscious subjectivity. But this is also not right. One’s capacity for being affected negatively by the world in specific ways is one’s vulnerability, one’s capacity to be hurt. Such vulnerability gives one a claim upon others that in pursuing their various interests they take this vulnerability into consideration. But one’s vulnerability does not constitute a carte blanche to demand that one’s sensibilities be respected just as they are and are understood by one oneself to be. Others are vulnerable in their own ways and these ways can conflict with one’s own such that to take the vulnerability of others into consideration is to deny one’s own and vice versa. If, therefore, one could only choose between subjugating one’s affectivity and letting it run riot, one could only show regard for others by disregarding oneself or, conversely, show regard for oneself by disregarding others.

Evidently, in order to avoid this invidious situation, a notion of affective response is needed which is ontologically rich and robust enough to allow for a sense in which one and the same affective response can be negotiated and re-negotiated in ethical discourse and experience, thereby assuming a form which represents a better balance between one’s own interests and those of others. Otherwise, we will find ourselves forced to construe distinctively ethical deliberation as merely ascertaining limits to purposive behaviour and ethical will as merely imposing limits. This represents a truncated conception of ethical deliberation and will, hence of rationality as such. Furthermore, since at least some degree of rationality is constitutive of self-consciousness, it represents a truncated conception of self-conscious subjectivity itself.

Once again, the case of Jane illustrates the point. At least as we have interpreted her story here, Jane does not, while in Spain, get habituated out of desiring an elegant or ‘cool’ appearance entirely. Rather, she reshapes this desire by reshaping her understanding of its goal or end, that is, her understanding of an elegant or ‘cool’ appearance. In Spain Jane learns not so much that this desire is wrong, but how, given that it is wrong in its current form, it may be re-formed so that it can get its due without thereby denying ethical considerations their due. From the outset, she understands her desire, hence the end or goal of this desire, to be susceptible to a distinction between different forms of realisation, different precisely with regard to how well it can realise itself without detriment to wider and more specific contextual considerations. Given the nature of its end or goal, this desire is inherently capable of different forms and contexts of realisation in which it is better or worse, more or less entitled to fulfilment.

Now if this is right, then something similar must apply to the underlying affective responsiveness. It must be inherently subject to a distinction between appropriate and inappropriate ways of giving it some due – more or less appropriate, given the current empirical and ethical circumstances. Jane’s sensitivity to the judgements and rejection of others is initially inappropriate, indeed ill formed , because it is exaggerated and insufficiently tempered by effective assessment of the opinions of others.17 In addition, it leads to wasteful behaviour which is unethical in the circumstances. But precisely because it comes, and is from the outset understood as coming, in this inadequate form, it points to the possibility of that other, thus far unelaborated form of itself in which it is legitimate, in which it is entitled to receive some due, both in Jane’s own practical deliberations and in those of others. So, too, with the disposition to respond with disgust and revulsion to body odour: it, too, must be inherently subject to a distinction between appropriate and inappropriate ways of giving it some due , more or less appropriate according to the particular empirical and ethical circumstances. It, too, inherently points to the possibility of its motivating, and being accommodated by, a mode of behaviour other than current one – other in that it more adequately acknowledges the claims of all parties. It therefore has an identity which transcends current ways of acting it out. Moreover, these different ways of acting it out are not just different, they are also better and worse, according to whether they represent more or less successful ways of accommodating the response without disregard to the empirical and ethical specifics of the situation.


In general, then, it must make sense to speak of a disposition to affective response as capable, not just of subjugation and eventual eradication, but of rational management in the course of which one renegotiates the ends and goals motivated by this affective responsiveness. Jane neither wages war on her past concern for an elegant, even ‘cool’ appearance, nor does she obstinately persist in it. Rather, in the face of recognition that her past practices of water use are untenable, she reworks her concern in the light of this knowledge – no doubt aided, of course, by the discipline imposed upon her by her new Spanish environment. In the course of this development, the underlying sensitivity to the judgements and rejection of others itself does not simply disappear; it changes and evolves for the better.18

But all this is only conceivable if one accords to affective response a certain ontological robustness: it must be capable of a certain sameness-in-behavioural-difference, that is, the same way of being affected by the world must be able to manifest itself one way, as one behavioural practice, and in another way, as different behavioural practice. Otherwise it would not be possible to negotiate and renegotiate desires, interests and ends over time and if this latter were not possible, then there could be no place in one’s account of self-conscious, more or less rational subjectivity and agency for the rational reconfiguration of desires, interests and ends.

Moreover, as the previous section has implicitly argued, this robustness must ultimately be founded in what affective response is a response to, namely, the objects of affective response. These must themselves display a certain sameness-in-difference across all the different ways subjects can relate behaviourally to them – a sameness which lies not in their character as satisfying certain affectively neutral, perhaps natural scientific descriptions, as body odour satisfies a description in terms of a certain mix of chemicals. Rather, this sameness must lie in their character as rendering certain characteristic affects intelligible. Thereby, the objects of affective response render intelligible any number of different behavioural practices instituted in order to deal with them. In this sense, then, people sometimes really do smell and behavioural practices can be better or worse in dealing with this.

Unfortunately, Shove’s tendency to constructivism engenders an eliminativism which prevents her from admitting such ontological robustness. In consequence, she finds it difficult to do justice to the kind of transition we are presuming Jane to have made. Shove does not have, I think, the conceptual resources needed if one is adequately to bring out the serious differences between this kind of transition and the kind of transition one makes when one moves from driving on the left hand to driving on the right hand side of the road. The latter is genuinely just a matter of re-habituation. But the kind of transition we are presuming Jane to have made is not. And only this latter kind of transition is of interest as far as achieving more sustainable everyday behavioural practices is concerned.

So in order for Shove’s work on everyday behavioural routines for securing comfort, cleanliness and convenience to yield deeper, more truly distinctive insights into how to promote Jane’s kind of transition, her constructivist tendencies must be corrected by the only kind of thing which could conceivably reveal the constitutive role of affectivity in rationality: an in-depth philosophical investigation of the various aspects of self-conscious agency. Such a philosophical investigation would constitute the necessary complement to the kind of empirical project undertaken by Shove; its capacity to correct overly constructivist tendencies would demonstrate its indispensability for the practical project of understanding better how to make transitions to more sustainable behavioural practice. This paper is a first step towards such a fundamental ontology, as Heidegger would call it.

Finally, let me note some further difficulties in Shove’s position which point to the need for a less constructivist, more robust notion of ends, desires and affectivities. Shove not only finds it difficult to accommodate the kind of transition we are presuming Jane to have made. It also prevents her from adequately interpreting some of her own data. Firstly, Shove refers to the culture of perfumery with which people in the seventeenth and eighteenth centuries sought to protect themselves against odours – see Shove 2003, pp.86-87. True, perfumes were then often employed as prophylactics against disease19 rather as masks for unpleasant odours since at the time it was commonly believed that disease was spread by vapours and gases. But even so, and whatever the reason for using perfume might have been, people were as sensitive as ever to foul stenches with which they were surrounded. Their affective response to odour was not simply a creature of initiation into the bathing and laundering practices of the day.

Secondly, Shove likes to emphasise diversity and difference in the practice and history of bathing and washing. This leads her, however, to downplay what cultures and traditions hold in common. The filth characteristic of Europe up until comparatively recently is arguably highly European: before, during and after the times when Europeans ostensibly despised washing and had to be dragged, like Louis XIVth, to their six-monthly bath, there have been numerous civilisations which, at least in their ruling classes, placed high value on bathing, washing and sanitation, e.g., the Greeks and Romans, the Islamic world from early on, the Mayans and so on. And even the reticence of Europeans to bathing was motivated by considerations of disease rather than a sheer visceral dislike of the bathing experience; as already indicated, at the time it was believed that washing weakened the body’s capacity to keep disease out. So refusal to wash does not signal an indifference to odour and dirt radically different from late modern affectivities.

Of course, in saying this and in arguing for the ontological robustness of affectivity and its objects, I am committed to defending what many would deny: that there really are improved standards, not just in the way individuals engage in practices, e.g., Jane’s improved water behaviour, but in the practices themselves. Shove shares in this tendency. At one point, she says that it makes sense to talk about ‘standards’ of cleanliness, i.e., standards internal to the practice. But in making this concession, she shows herself far less prepared to regard these practice-internal standards as ‘improvable’, i.e., as orderable according to better or worse, which would permit an ordering of practices of cleanliness: “Although it … makes sense,” she says, “to talk about ‘standards’ of cleanliness, and although some would claim that such standards have ‘risen’ historically and converged socially, the mechanisms involved differ from those associated with the management of the indoor climate.” (Shove 2003, p.93) Note the insinuation here that she does not belong to those who regard standards of cleanliness as having risen.

That there has been such a rise seems to me, however, to be undeniable. Nor does anything adduced by Shove really show the need to deny it. This brings us back to an issue raised above: what Shove is really getting at when she says that for participants in contemporary laundering practices, “anything that had been through the washing machine was, by definition, clean.” (Shove 2003, p.147) It makes no sense to regard this kind of claim as fixing the very meaning of the word ‘clean’. What, then, were manufacturers of washing machines really doing when, as Shove maintains,20 they encouraged consumers to think that anything which had been through the washing machine was clean? They were not in fact re-defining the word ‘clean’ or the concept of cleanliness. Rather, they were providing launderers with new (and ostensibly better) criteria for determining, naturally only ever fallibly, whether something was clean or not. Fallible criteria for the application of a word or concept presuppose, they do not constitute the meaning of the same. So the manufacturers must be understood as providing new criteria of application for an old word or concept. They were using the very same concept of cleanliness as had been operative in the days before washing machines.


Anscombe, G.E.M. 1956 Intention, Oxford: Oxford University Press

Shove, E. 2003 Comfort, Cleanliness and Convenience – The Social Organisation of Normality, Oxford and New York: Berg

Heidegger, M. 1979 Sein und Zeit, 15.te Auflage, Tübingen: Max Niemeyer Verlag; translated by J. Macquarrie and E. Robinson as Being and Time, Oxford: Basil Blackwell, 1962

Shove, E. 2008 “Urban Environments and Sustainability: thinking about Transitions in Practice”, Notes for a talk in Paris, October 2008

Shove, E. and Walker, G. 2007 “Commentary”, in Environment and Planning, Vol. 39, pp.763-770


  1. In Shove 2008, p.4.

  2. In Shove 2008, p.4.

  3. Shove might regard this claim as confirmed by the case of Elizabeth Drinker, wife of a well-to-do Quaker merchant, who described how well she ‘bore’ the experience of her first shower (in 1798), noting she had not been ‘wett all over att once, for 28 years past’ – see Shove, p.93. This kind of case shows, or so one might argue, that regular showering is not an intrinsically attractive or desirable activity; one has first to acquire the habit of regular showering in order then to find in it something attractive, even essential.

  4. On p.133 of Shove 2003, Shove interpolates a further thought, namely, that washing machine manufacturers and other commercial interests, through their role in provisioning practices of laundering, are able to determine how cleanliness and laundering get defined within these practices.

  5. Note that it does not help to weaken this to the claim that necessarily in order to get things really or truly clean, one has to put them through the washing machine. For even real or true cleanliness, whatever this might be precisely, is not, and would be acknowledged by Shove’s interviewees not to be, only achievable by a washing machine.

  6. Obviously, now that she is back in America, the prudential implications of intensive water use will have gone because in America she can get away with things she was not able to in Spain.

  7. In Anscombe 1956.

  8. And this identification can come about as a result of our being open enough to learn how to participate in the relevant form of life.

  9. All intentionality, hence rationality and interpretation presupposes affectivity or what Martin Heidegger calls Befindlichkeit – see Heidegger 1979, § 29.

  10. See Shove 2003, Ch. 2, esp. pp.28-34.

  11. This is known in Australia as the ‘yuk’ factor. How serious an obstacle it is to the introduction of potable re-use of water is unclear.

  12. One might rather sloganistically describe this as an eliminativism about comfort, cleanliness and convenience.

  13. This is not to deny, of course, that the initial affective response will, in addition and simultaneously, cause all sorts of non-intentional, ‘instinctive’ behaviour, e.g., gasping for air, holding one’s fingers to one’s nose, or perhaps even a flush of embarrassment at the awkward situation into which one has been put, etc.

  14. At least for the case of human beings. This reflects recognition of a standard functionalist point, as applied to the relation between body odour and the mix of chemicals which no doubt materially suffice for it in the case of human beings: there can be no type-type identity here.

  15. To argue that such eliminativism is false is not, of course, to say anything particularly substantive about the connection between natural scientific and everyday descriptions of phenomena such as body odour. In principle, it would be open to one to embrace an ‘Aristotelian’ conception of the connection as one of material sufficiency; or a conception of it as type-type identity; or finally a conception of it as token-token identity. Clearly, a conception of the connection as type-type identity founders on an adaptation of the standard functionalist objection to type-type identity in the philosophy of mind. So the only real contenders are the first and the third. I am inclined to opt for the first – a particularly radical choice, given the current ideological hegemony of naturalism. But the third seems to me forced to adopt some kind of purely functionalist account of odours, colours, sounds, etc. – the so-called secondary qualities. Then, however, it is incapable of characterising the character of the perception of odour, colour and sound in that sensual character which underpins and makes sense of affective response to odour, colour and sound.

  16. This is indirectly confirmed by something to which Shove herself refers, namely, the culture of perfume with which people in the seventeenth and eighteenth centuries sought to protect themselves against foul odours – see Shove 2003, pp.86-87, and below.

  17. Obviously, the causes of these deformations will be such things as teenage insecurity and lack of skill in assessing the judgements and opinions of others.

  18. And of course in consequence she changes for the better – better in the sense of becoming more autonomous in her self-assessment and assessment of others.

  19. See Shove 2003, p.86.

  20. See Shove 2003, p.133.