PUBLIC TRANSPORT SYSTEMS: THE SINEWS OF EUROPEAN URBAN CITIZENSHIP? Open Access

Public transport is often given a key role in the reinvigoration of European cities. The paper explores the relationship between urban citizenship and urban transport through a study of four cities (Athens, Bologna, Dublin and Helsinki). Urban citizenship is considered to comprise both ‘social cohesion’, i.e., the participation in the public space of the city, and ‘social inclusion’, the right of all inhabitants to physical mobility within the city and hence to access employment and social facilities. Despite the claims of urban planners, there is only weak evidence that reducing car usage contributes to social cohesion, but there is however stronger evidence that reducing car dependency contributes to social inclusion.

In European cities today, transport policy is often at the centre of political debate. Amongst architects and planners it is usually axiomatic that cars and urban living are antithetical: reviving Europe's cities involves curbing the car and developing public transport.

Sometimes explicitly, sometimes implicitly, such policies call for a reinvigoration of urban citizenship. This paper argues that the concept of urban citizenship can be developed analytically by focusing on two different dimensions, termed here social cohesion and social inclusion. Public transport can potentially contribute to social cohesion by creating common public spaces, and to social inclusion by enabling all inhabitants of the city to move around the city. Conversely, high car usage undermines social cohesion and high car dependency exacerbates the risk of social exclusion.

Using a study of four European cities, the paper empirically investigates these arguments. After an exploration of the possible relationships between urban citizenship, public transport and the private car, the second part of the paper shows that while Athens and Dublin appear to be car dependent, Bologna and Helsinki have restrained the private car and developed public transport. The third part of the paper shows that in these cities there is some limited evidence that high car usage undermines what has been defined here as social cohesion. However, there is a much stronger case that car-dependent cities exacerbate social exclusion. The fourth part of the paper compares two working class suburban areas in Dublin and Helsinki, showing that that effective public transport, and hence low car dependency, can contribute to social inclusion. In that sense, public transport can hold a city together, contributing to developed urban citizenship.

Recently it has become common to speak of a resurgence of the European city and even of urban citizenship (Delanty 2000). According to this perspective, the power of the nation state has moved both ‘upwards’ to the European Union and ‘downwards’ to the city. The new Europe is in part a Europe of cities (Le Galès 2002: 145).

One justification for this view is the emergence of high profile revival strategies in some cities, often linked to high profile urban politicians: Pasqual Maragall in Barcelona, Pierre Mauroy in Lille, even Ken Livingstone in London. Such endeavours often involve some cultural ‘rebranding’ of the city, as famously in the transformation of Glasgow (Warhurst et al. 2000). They also involve prestige ‘signature’ buildings (most recently, the Guggenheim in Bilbao) and – the concern of this paper – the re-construction of urban transport systems, spearheaded by the revival of the tram in Strasbourg under Catherine Trautmann.

Both scholars and politicians appeal to the golden age of the European city at the end of the middle ages, yet this was only a brief interlude before the consolidation of the nation state (Le Galès 2002). In fact the European city had a more recent and relevant ‘resurgence’. In the nineteenth century Britain developed almost ex nihilo a new tradition of expansionist urban government which contributed to the development of urban transport. While this declined in the twentieth century (Hunt 2004), a similar tradition flowered briefly in Weimar Germany (e.g., Frankfurt am Main under Landmann) (Rebentisch 1975) and in the Vienna of pre-Anschluss Austria.

Civic revival raises the question of citizenship. It is the ‘citizens’ who are defined as the primary beneficiaries of the urban, it is the ‘citizens’ who are summoned to actively ‘Build Jerusalem’. Urban citizenship however is rather vague. From a legal perspective the notion of urban citizenship must be problematic: unlike the city of the late middle ages, the contemporary European city cannot confer any legal status. Local government often administers many social services and so has some limited leeway in determining the scale of provision, but access to them remains determined by national citizenship status.

Nonetheless, urban citizenship does have meaning as part of the public realm or even the ‘public’ itself. According to Marquand (2004), the ‘public domain’ is that ‘dimension’ of social life which is neither the market nor the private. From this perspective, one role of the city is to provide physical spaces where a public life can be enacted, whether these be the grand ‘public’ spaces (squares, plazas, etc.) used for demonstration and symbolic events or the smaller scale ‘parochial’ meeting places of playgrounds and parish halls (Beauregard and Bounds 2000). The city provides a physical ‘agora’ or meeting place. Echoing the civic life of the European mediaeval city, this is a local polity to which anyone can have access, a polity that operates on an on urban scale and in urban places. Urban citizenship is thus one physical aspect of the public realm.

Given the decline in public engagement (Putnam 2000), the image of a city in which citizens are active from the neighbourhood residents’ association through to the grand public demonstration is more aspiration than reality. However, people who live in the same city, even if they do not live in the same neighbourhood, do share common physical places and common physical spaces. This common use of shared physical spaces I term the ‘social cohesion’ dimension of urban citizenship. Social cohesion indicates that, despite differences, citizens share a common public life. Participating in such life is to share public spaces with other strangers: the public realm thus depends on ‘civility’ (Sennett 2002). A lack of social cohesion, following Durkheim, involves anomie, a lack of trust, an exaggerated individualism, and in practical terms, the lack of such public spaces.

A long tradition in architecture and town planning has been concerned with the city as public space. This tradition positions itself against those functionalists and modernists, from Ebenezer Howard (founder of the English Garden City movement) to Le Corbusier, for all of whom the ideal was a city of functionally separated spatial zones. By contrast the ‘urbanist’ tradition within the Anglophone world extols diversity of usage within urban areas. The key text here is Jane Jacobs’ The Life and Death of Great American Cities, a passionate defence of diversity of usage and population in city ‘neighborhoods’ (Jacobs 1962). For Jacobs such diversity ensures that pavements (‘sidewalks’) are continually used by a wide range of people, some of whom live or work in the local area. In her famous phrase, there are therefore ‘eyes on the street’ (Jacobs 1962: 35). Such ‘vie de quartier’ ensures that the streets are safe, strangers are accommodated and children can be watched over and so assimilated to adulthood.

We can suggest that transport technology has several implications for this form of social cohesion. Firstly, both suburban transit and later the automobile enabled functions to be separated in space. Although the car is now held responsible for such dispersion, it was mass transit that created the suburbs of nineteenth-century British and US cities. Thus Los Angeles became the most suburbanised city in the USA not because of the car but because of its extensive electric rail system (Bratzel 1995). For Putnam (2000: 210) the key issue is spatial segregation, not just of home, work and shopping, but also of residential areas, each inhabited by a narrowly defined social category (see also Duany et al. 2000). In the same vein a leading British architect (Rogers 1997) argues that dispersed cities are not only environmentally unsustainable because of their high energy usage, they are also less socially cohesive, separating functions which in the urban neighbourhood occur within the same physical area.

Secondly, car transport involves the physical destruction of public space. Jacobs campaigned against the insidious road widening which narrowed pavements in New York City; many American suburbs are now built without any pavements at all. In a car dominated environment space becomes single use. The credo of the architectural movement loosely known as ‘the new urbanism’ (Katz 1994; Duany et al. 2000) is that multi-functional space contributes to civility. Thus making a point continually stressed by Jacobs, Rogers argues that ‘open-minded’ spaces are ‘multi-functional’. The car park, the shopping mall, the motorway, all serve only one function. By contrast public spaces (‘the busy square, the lively street, the market, the park, the pavement café’) serve many functions: they are places where ‘we are readier to meet people's gaze and participate’ (Rogers 1997: 9).

Thirdly, though this is not often explored in the literature, public transport can be seen as itself public space. Users of public transport have to share physical space with others, as opposed to being ‘carcooned’ alone in their metal box. Today, when SUVs are ubiquitous and derivatives of the US Army armoured Humvee are sold as private cars in the USA, Jacobs was remarkably prescient: she described one response to unsafe streets as people ‘hiding in their cars … like tourists in the game reserves of Africa’ (Jacobs 1962: 46). Users of public transport do not have this option, they are compelled to share space with strangers.

Of course, car drivers may find the mere fact of being alone in the car pleasurable – a peaceful space between the sociability of home and workplace, and this is facilitated by the improvement of in-car entertainment systems (Putnam 2000: 213). The car can also be seen as an extension of the person, so that the car-and-driver form a hybrid ‘actant’ (Urry 2002). This makes it easier to see car-driving as learning civility and the decline of overall car accidents as a part of the social internalisation of rules – Elias’ civilising process (Elias 1995). From this counter-perspective then, the car-and-driver also participate in the public space of the highway.

Finally, public transport can be seen as ‘public’ in the sense that it is publicly owned or regulated. At the grandest level, European public transport is part of the enhanced public realm that distinguishes Europe from the USA:

Again there is an obvious counter-argument: roads are usually publicly owned and part of the state, as in the English expression ‘the Queen's highway’. Even when roads are built and operated by private companies, their construction requires political decisions and their use public regulation. Claims about the inherently ‘public’ nature of public transport require empirical investigation.

Urban citizenship can also be considered to involve social inclusion, namely the extent to which all inhabitants are able to participate in the ‘normal’ activities of the city. Transport enables inhabitants to physically access such activities despite separation in space.

From Henry Ford on, the great appeal of the car has been its democratic nature. The car was democracy on wheels. In the epoch of the family car, the father-worker had priority access over the (house)wife. Even today, it is usually the case that women use public transport more than men, and in older age cohorts women are less likely than men to have a driving licence. However, the obvious solution to this gender inequality problem was – more cars: one for him and one for her. First in the USA, and then in Europe, the family car has been partially replaced by the individual car. Already by the 1950s the normal US suburban family had two cars, and in the early 1960s Jacobs was reporting that the suburban housewife's car travel for school runs and shopping meant that she drove further than her husband on his simple commute to and from work.

Once high levels of car usage occur, a society (or a particular city or area) can become car dependent. Car dependency means that a car is necessary in order to participate in the normal activities of the society – to work, shop, socialise, etc. Such ‘activity spaces’ (Hine 2003) are too far away to reach on foot, while other forms of transport have either withered or never existed in the first place.

Car ownership today is so widespread that it is often forgotten that many adults do not own their own car and must share with other members of their household. Some people (children, young teenagers) are too young to drive a car, while in an aging population there are increasing numbers of older people for whom driving is dangerous or frightening. And finally, there remain adults who simply cannot afford a car.

Only hesitantly is transport policy becoming connected to social policy (Cahill 1994; Social Exclusion Unit 2003). Thus, there is some evidence that those without cars are excluded from employment opportunities. While in the USA this is particularly a problem for inner city residents unable to reach new employment in the suburbs (see Shen 1998), in the UK it is more a problem of the difficulties of suburb to suburb commuting for those without cars (Webster 1999). Yet such analysis remains fragmented: we have no systematic exploration of the relationship between car dependency as a whole and social exclusion.

In areas without public transport and where activity spaces are physically dispersed, transport exclusion can be either ‘clustered’ or ‘scattered’ (Hine and Grieco 2003). The former is particularly a question of low income areas, where many are too poor to be able to afford a car, while the latter is more a question of those who cannot drive because they are in some sense not allowed to (the old and the young). Once car-based activities have become normal, then those who do not have access to a car are ‘poor’ or ‘excluded’:

This famous definition of relative poverty alerts us to the implications of not owning a car in a car-dependent society. An anti-poverty strategy should therefore include transport policy, and conversely, transport policy itself needs ‘poverty proofing’ (Social Exclusion Unit 2003). Evaluating the contribution of public transportation to social inclusion involves evaluating the extent to which all citizens can move around the city. Such a right is even more to difficult to define (‘how much mobility at what cost?’) but the principle is clear: a right to mobility depends on the availability of transport apart from the private car.

The extent of such a generalised right to mobility is not fully measured by conventional social cost benefit analysis which determines the costs and benefits of transport provision for different social groups. Examining such analysis in the USA, Hodge (1995) has argued that a focus on the equity of transport provision on specific routes may well detract from the general benefits of public transport which are met when there is an effective public transport system. The technical requirements are obvious enough: routes which are not simply radial but orbital and even grid-like, an integrated network so that transfer from one mode to another is easy (modal interchanges, integrated ticketing), effective and comprehensible information that allows facilitates general mobility.

Since cities vary in the extent to which they have such transport systems, these arguments suggest that, even given broadly similar levels of car ownership, cities will also vary in the extent of their car dependency. Where there is an extensive transport system, and/or activity sites are physically close to one another, those without a car will still be able to participate in the ‘normal’ activities of the society. Conversely, where these conditions do not occur, such individuals will be excluded from society, and car ownership becomes an entry condition for full citizenship.

The previous section has suggested a certain technological determinism: urban citizenship is influenced by the physical layout of the city and the extent of its public transport system. In cities with high car usage, social cohesion will be undermined; in cities which are also highly car dependent, social exclusion will be exacerbated. This section of the paper now briefly outlines the research used to investigate these arguments.

The paper derives from a study² of urban transport and mobility in four European cities. Whereas much literature in this area involves case studies of best practice (e.g., Apel and Pharaoh 1995), the project contrasted two apparently worst practice cities (Athens and Dublin) with two relatively successful cities (Bologna and Helsinki).

The project investigated the mobility of different groups of people in different areas of different cities. The study began by comparing the four case study cities with other cities across the globe in terms of the overall level of car usage.³ Using such data from 1990, other authors (Kenworthy et al. 1997) have already shown how car usage is significantly lower in European rather than American cities. Adding our case study cities to the data set showed how car usage was higher in Dublin and Athens than in Helsinki and Bologna. Since at the time Helsinki and Bologna were the richest of the four cities, this cast some doubt on the easy assumption that car usage must rise with GDP. Even more interestingly, the data challenged the usual belief that low density cities necessarily have high car usage. Both Dublin and Helsinki are low density cities, but Dublin's car usage is far higher than Helsinki's.

To explain such differences the project involved a social historical study of the evolution of transport in the four cities, as well as an analysis of key political transport decisions in each city. In Bologna and Helsinki the increasing dominance of the car was challenged in the 1960s and 1970s by policies which developed integrated public transport systems and to some extent restrained car usage. This historical turning point launched both cities onto a particular development trajectory supported by an urban coalition of diverse interests. By contrast Dublin and Athens made little or no investment in public transport, and road building was restrained only by bureaucratic inertia rather than any political choices. This historical section of the research thus modifies the technological determinism of the research hypothesis. While car usage and car dependency may have social consequences, the first stage of our study showed that their extent varies depending on socio-historical decisions.

The main empirical work of the project however ‘zoomed in’ on areas within each case study city. In each case study city three distinct urban areas were selected, broadly characterised as suburban working class, suburban middle class, and inner city professional (‘yuppie’). The total project therefore collected data on 12 areas presented in Table 1.

Research in the locality enabled investigation of transport usage in specific spatial contexts. By locating the availability of facilities (shops, entertainment, etc.) it linked questions of mobility (how people move around) with questions of accessibility (what people can reach). Apart from collecting any existing secondary material (research reports, population statistics, pollution data), fieldwork involved visiting the area and driving and walking around, as well as using public transport; interviewing key informants, ranging from town planners to local taxi drivers; and taking photographs.

The research continued with a questionnaire-based social survey in each area. Since this stage of the study aimed to compare different areas of the case study cities, the sample was designed to be representative of each individual area, not of the city as a whole. Each case study area was divided into two or three further sub-areas, within which the sample was selected by a random walk, with equal quotas for women and men, and a total of 50 respondents in each area. In Bologna it was possible to increase the sample to 100 respondents in each area.

The survey showed that the aggregate differences in car usage used to contextualise the four cities remained valid nearly ten years later and at local level. Table 1 shows that at household level car ownership is normal in all the areas, even if it is lower in the inner city. Household car ownership is lowest in inner city Helskinki and highest in middle class suburban Athens. Furthermore, there are significant differences between the cities, as well as just between the areas, in terms of public transport usage. Public transport is used more in Helsinki and Bologna than in the other two cities. Particularly dramatic is inner city Helsinki, where 58 per cent of all respondents (and even 43 per cent of all with access to a car) had used one or more of bus, tram or metro within the last two days. By contrast in inner city Dublin, only 20 per cent of respondents had used public transport, and these were exclusively those who did not have access to a private car (Table 2). These usage levels are broadly in line with satisfaction with public transport as measured by a series of questions in the survey: satisfaction is significantly higher in all three areas of Helsinki than in any other area of any other city.

Empirical work in the areas concluded with focus groups with local residents. The meetings started with the participants drawing a map of their journeys of the previous day, the maps serving both as starting point for the focus group discussions and providing data about patterns of mobility. All fieldwork was carried out during 1999.

The case study cities do show very different forms of public space. Bologna is famously a city in which the piazzas (above all the Piazza Maggiore itself) and the covered arcades do function as ‘outdoor living rooms’. Helsinki, and to a lesser extent central Dublin, also has city centre public spaces. At the other extreme, most of central Athens is unpleasant with few public spaces that are not over-run by private cars. Such issues of urban design involve the extent to which city centres are spaces for strolling, meeting, flaneuring, rather than just transiting from one destination to another.

Given the argument that curbing traffic restores the streets to pedestrians and recreates public space, the extent to which people wish to walk around the area in which they live is an indicator of social cohesion. The survey therefore asked a series of questions about the ‘walking quality’ of the area, covering aspects of traffic pollution, personal safety and the safety of children (Appendix). The results show there is no straightforward relationship between sociability and personal safety. Thus the measure of sociability (‘If I walk around this area I nearly always meet someone I know’) does not correlate at all with the measure of perceived personal safety (‘I personally feel safe from any physical attack (e.g., mugging) while walking around this area’). People feel safest in the middle class suburban areas and were most likely to disagree with the statement ‘I personally feel safe from attack …’ in the working class area suburban of Dublin (60 per cent) and Athens (55 per cent), and in inner city Bologna (58 per cent). This suggests that the advocacy by architects such as Rogers of pedestrianised streets is naïve. Indeed, there is growing anecdotal evidence that pedestrianisation of inner city areas attracts ‘street people’ whom the inhabitants find threatening (shortly after the survey was carried out, there was a much publicised murder in the Piazza Maggiore in the heart of Bologna's Centro Storico). Equally, the Dublin working class suburban area was hardly perceived as unsafe because the car had driven away pedestrians.

The different items on walking quality were combined into a simple additive scale. The single item concerned with sociability (‘If I walk around this area I nearly always meet someone I know’) did not correlate strongly with the others and was accordingly dropped from the scale. The resulting scale based on seven items had an alpha of 0.79 which was considered acceptable. The box plot (Figure 1) shows how walking quality is higher in all Helsinki areas than in comparable areas in other cities. It is noticeable that walking around inner city areas is not seen as particularly pleasant by their (largely middle class) inhabitants. Indeed, inner city Bologna, praised by planners because of its restrictions on car usage, actually scores the lowest of all areas in terms of walking quality. Equally surprising perhaps is that there is little difference between the genders here and living in a car owning household also makes very little difference.

The figure also shows the high ‘walking quality’ of all three Helsinki areas: the working and middle class suburban areas (Kontula and Länsi-Pakila) have the two highest median scores of all 12 areas; the inner city area (Taka-Töölö) has the highest median score of all inner city areas. Working class and middle class suburbs in Dublin and Helsinki appear very similar: all have plentiful public space, mostly in the form of open areas between buildings. Yet the ethnographic reports show a clear contrast. In Clonskeagh what strikes one, according to our ethnographic reports, is ‘the quietness of it all’, whereas in Länsi-Pakila ‘there are people walking – kids and adults’. One reason for this is simple: in Länsi-Pakila, although car ownership is high, most children go to school on foot, by cycle or by public transport; in Clonskeagh most are driven to school by their parents. The physical layout of Länsi-Pakila facilitates cycling and walking in a way that Clonskeagh does not. In Clonskeagh the main roads around the area frequently have no cycle lanes or even footpaths, thus discouraging cycling and walking outside the immediate estates, although within the estates there are adequate footpaths and traffic is calmed by speed ramps. In Länsi-Pakila by contrast roads around the area, even though busy, have well kept pedestrian/bicycle lanes. Within the area there are often pedestrian/bicycle gravel lanes which traverse the area almost completely within woods. Even in winter many are swept clean and all are in use. Accordingly, in Länsi-Pakila the physical layout facilitates those who wish to walk or cycle, thus creating small scale public mobility spaces, whereas in Clonskeagh, despite the milder climate, the car is the only mode of transport which is facilitated and public space becomes ‘dead space’ (Jacobs 1962: 90).

The data suggest that public transport itself is to some extent also seen as public space. Across the study as a whole those who travelled to work by public transport were marginally more likely to enjoy their journey than those of travelled by private car; there was an almost complete consensus that ‘The time is past when one could enjoy driving the car [in the city]’ (focus group respondent, Helsinki). In Bologna and Helsinki, some focus group participants described travelling by public transport as pleasurable in itself. For older residents in Bologna the efficient bus service was a source of social contact. As one remarked: ‘At our age, the more people you see the better’. For such respondents, public transport was pleasurable because it involved shared public space.

However, once again, this public space can be destroyed. Sociological discussion of citizenship has spent more time on the ‘rights’ of citizenship than on its ‘duties’, yet legal theory shows that rights necessarily involve restraints on the rights of others (Pagano 2001). Thus the right to public space imposes the duty of appropriate public behaviour towards others in that space. The only way to avoid this duty is not to exercise the right – and to withdraw to private space (Beauregard and Bounds 2000). Loutish and threatening behaviour, graffiti, littering, all indicate that the perpetrator exercises his or her use of the facility at the cost of others. In the study this problem was almost entirely restricted to Dublin. Graffiti, littering and threatening behaviour are common on some routes; bus drivers have several times gone on strike because of the violence they face from passengers on buses serving some working class areas. One respondent recounts:

Such behaviour has of course a simple consequence – other citizens abandon the facility as soon as they can, usually for private transport. If public transport is to create public spaces and citizens exercise their right to mobility, then the duty of public civility has to be enforced.

In Europe car ownership is widespread, but not universal even at the household level. Thus, household car ownership in our 12 areas ranged from 90 per cent of all households in middle class Athens and 89 per cent in middle class Dublin to, at the other extreme, 39 per cent in inner city Dublin (Table 1). If the analysis is restricted to those households which comprise two adults both in their middle years (‘core age partner households’), household car ownership becomes almost universal, only dropping below two-thirds of all households in working class Dublin (63 per cent). In middle class Dublin car owning households, there are 0.59 cars for every adult household member, whereas in working class Dublin households the figure is only 0.30.

Extensive car ownership or even car usage does not in itself prove car dependency. When it is claimed that Belfast is the UK's ‘most car-dependent city’ (Cooper et al.2001), this is strictly speaking unproven, since it is merely deduced from the fact that car usage is higher in Belfast than elsewhere. The simplest aspect of car dependency is that a particular journey can only be made by car – the journey is mode dependent. Of course, the extent to which there is no ‘choice’ can vary between individuals. For example, there may be public transport available, but it may take longer and some people may feel they ‘cannot’ spare the time. It is equally plausible that car users may become simply unaware that there is a public transport alternative; they may under-estimate the quality of the service, etc. These issues apply to each specific journey, but evaluating ‘car dependency’ must consider different journeys. For example, for many people in Europe work and home are linked by a suburb to city centre transport corridor which contains both roads and mass transit, so that it is meaningful to talk of ‘choice’. By contrast, at the weekend these same people may shop in a supermarket in another suburb which they can only reach by car. Here ‘choice’ would mean a different destination altogether.

So long as ‘car dependency’ is conceptualised in terms of existing journeys it is relatively simple to measure. In the survey respondents were asked how they travelled to work and to other non-work activities (shopping, using medical facilities, visiting friends, entertainment); for each activity they were also asked whether there was another form of travel they could ‘reasonably’ use. For the journey to work this gave a measure of car usage (the proportion of those at work who travelled to work by car) and of car dependency (the proportion of those car journeys for which respondents claimed there was no alternative to the car). Responses for all the non-work activities were aggregated to give a similar measure of car usage and car dependency for non-work activities.

Before proceeding to the empirical analysis it is important to notice that such measures of car dependency only begin to tackle the question of car-dependent lifestyle. Firstly, they only report dependency at one point in time. They do not explore if and how car usage drives out other forms of mobility, in particular through its connection to changing spatial patterns of housing, work and sociability. Secondly, they conceptualise mobility as a series of discrete journeys between fixed points, each for a specific purpose. Our ethnographic reports show that while this understanding of mobility is typical of car-users, some city dwellers, especially those who walk and use public transport, see mobility more in terms of ‘moving about’, ‘about being in contact with one's surroundings and with other people’ (Rajanti 2002: 119).

With these reservations in mind, Table 3 shows the ‘normal’ mode of travel to work for the core-age employed group: at one extreme, 76 per cent of respondents in middle class Dublin normally travel to work in their own car, at the other extreme the figure falls to 9 per cent in inner city Helsinki.

Comparing car usage and car dependency usage shows that much car usage for the journey to work is a matter of choice. In middle class Dublin only 14 per cent of all journeys to work are car dependent and only 18 per cent of all who drove to work claimed that they had no ‘reasonable’ alternative. Indeed, there are only two areas (middle class Athens and inner city Helsinki) where a majority of car users considered that they had no alternative to the car for their journey to work.

These results show that the proportion of car journeys that are mode dependent does not vary directly with the level of car usage. Thus the area with the highest proportion of car journeys for which there is no alternative is inner city Helsinki, where virtually nobody uses a car to travel to work (recall that here car usage for this journey is an astonishingly low 9 per cent). However, absolutely everyone who does use a car, considers that they have no option (car dependency for this journey for car users is 100 per cent). In inner city Helsinki, the ethnographic reports show that those who use their car for work not only travel out of the city centre, but also are travelling and working at the same time (for example, an architect visiting a series of clients in one day). Thus the low level of car usage in inner city Helsinki can be seen as the irreducible minimum that occurs when re-urbanisation is combined with a highly effective and well regarded transport system.

By contrast, the next highest proportion of mode-dependent car journeys is in middle class Athens (74 per cent). This is an area where half of all journeys to work are by car, one of the highest levels of car usage in the study. The high proportion of mode dependency amongst car users in middle class Athens is the result of a public transport system that is so bad that very few could consider using it and an urban environment in which walking or cycling to work is unacceptable.

The contrast between inner city Helsinki and suburban middle class Athens suggests that for car owners the relationship between overall levels of car usage and car dependency may be curvilinear. Where few people use cars, car dependency for those who do use a car is high: people only use a car when it is absolutely necessary to do so. As car usage rises, this is initially because the car becomes preferred to other forms of transport, which however still remain available. Finally, as car usage continues to rise, other modes of transport decay and increasingly the city is built on the assumption that everyone does have a car: car usage becomes ‘essential’ and car dependency for car owners becomes complete.

Figures 2 and 3 investigate this argument. Using the overall results for each area, Figure 2 plots travel to work car dependency (vertical axis) against travel to work car usage (horizontal axis). It shows that in fact for this journey car dependency falls as car usage rises (the two items have a correlation of –0.57. This confirms what we have already seen: in most of our city areas people both have access to a car and could use other forms of transport to travel to work. The figure shows that middle class Athens is the only case with high car usage and high car dependency, so that in our other case study areas car dependency has not yet (?) reached American levels, where there is literally no alternative but the car for most journeys to work. If however public transport atrophies and suburbanisation continues, we would expect car dependency to begin to increase with rising car usage.

Figure 3 shows that for non-work journeys car dependency increases as car usage rises (the two items have a correlation of 0.46). In fact here the relationship is curvilinear since the figure shows three clusters of areas: one of low car usage and medium car dependency (Helsinki Inner and Working, Dublin Inner and Working, Athens Inner); one of medium car usage and low car dependency (Bologna Middle and Working, Helsinki Middle); a final cluster of high car usage and high car dependency (Athens Working and Middle, Dublin Middle). In other words, for non-work journeys car dependency at first does fall as car usage rises: as more people use cars, they make journeys for which there remain alternative modes of transport. However, as car usage continues to increase, more and more journeys are made for which the car is the only possible form of transport.

Figure 3 also shows how the relationship varies by city. In Bologna and Helsinki, as car usage rises, so car dependency falls. By contrast, in Athens and Dublin, as car usage rises car dependency increases.⁴ This is consistent with the better quality public transport system in Bologna and Helsinki as compared to the other two cities: people are choosing to make journeys by car, but there are still public transport alternatives even for many non-work journeys. By contrast in Athens and Dublin, increasing car usage means that people are making more journeys (e.g., suburb-to-suburb) for which there is no public transport alternative.

If ‘normal’ life in a city depends on having a car, then obviously those without cars run the risk of being excluded from normal life. Thus the high car dependency of a small number of individuals, as in Taka-Töölö (Helsinki Inner) is not a problem, for most people are able to use other forms of transport (overall car usage is low). Equally, where car dependency is high and car usage is high, as in Polidrosso (Athens Middle), social exclusion is not a direct problem because nearly everyone is able to use a car. Social exclusion is potentially a problem at the intermediate level of dependency, where a substantial number of people travel by car and some of them with a car are able to access opportunities which have both become defined as ‘normal’ and remain denied to the significant minority without a car.

Once the car becomes essential for a significant number of people, we can assume that non-car owners are going to feel under great pressure to become car owners. In innovation studies such patterns of technological dependency are known as ‘network effects’. As usage of a technology increases, so the cost of not using it rises, and those who refuse to do so risk social or perhaps professional marginalisation (Dupy 1999). In the case of transport, this can involve social exclusion. For many, car ownership may be the only way they can access reasonable jobs. Achieving car ownership may impose financial strain on the household, while not achieving it means household members are disadvantaged – in ways that do not occur in less car-dependent areas.

The extent to which the car becomes a ‘necessity’ in this way will depend in part upon the quality of the public transport system. As we have already seen, in these terms Helsinki is clearly superior to Dublin. We now examine the extent to which public transport reduces the risk of social exclusion by comparing the two working class suburban areas of Kontula in Helsinki and Jobstown in Dublin.⁵ Within the study Jobstown is the most clearly working class area: in terms of the standard classification used in the survey,⁶ over 80 per cent of respondents were ‘working class’, the highest in the study; in 1996 male unemployment was 27 per cent. While in Kontula only 47 per cent of respondents were classified as working class, at the time of the survey unemployment was 22 per cent, nearly the highest of all areas of Helsinki. Both are areas of public housing built for a low income population: Kontula in the late 1960s, Jobstown in the late 1980s. In Kontula housing is concrete apartment blocks, in Jobstown low-rise terraced housing. Their building styles and their peripheral locations are typical of post World War II working class housing in their respective countries (Power 1998).

Dublin as a whole lacks any integrated public transport system, and services in Jobstown are minimal. One bus route links the area to the nearby Tallaght Town Centre which is the nearest large shopping facility 1.6 km away; the bus also links to Dublin City Centre, itself several kilometres away. The bus shelters are often vandalised and indeed on occasions the service has been withdrawn because bus drivers have been attacked by local youths, and there are few local facilities within walking distance. Given the low level of car ownership, it is therefore likely that those without access face increased risks of social exclusion.

In Jobstown the car is seen as a necessity for access to employment. In the survey car owners were asked what problems would arise if they did not have a car; non-owners were asked the benefits they could imagine from having a car. In Jobstown 44 per cent of owners mentioned employability as a potential problem, and 33 per cent of non-owners mentioned employability – the highest area in the study.⁷ The importance of the car in this area is shown by the fact that had the highest level of travel to work by private car amongst car owners, even though it had one of the lowest levels of overall car ownership in the study. Having a car, in other words, enables people to access a greater range of employment: car ownership enables a qualitative shift in employment opportunities. In less car-dependent areas, those with a car may choose not to use it to travel to work, since employment opportunities are not car dependent.

A quote from one of our interviews⁸ documents what is involved, as well as showing the financial pressure of car ownership in low income households:

For men in Jobstown workplaces are normally the new factories and logistics centres along the M50 orbital motorway. These are relatively easy to reach by car from Jobstown, but using public transport would have involved taking a slow bus into the city centre and then another radial route out again.

For those without a car, shopping for large amounts of food involves walking or going by bus to Tallaght Town Centre, and perhaps pushing a supermarket trolley back along the road. There are few good leisure facilities within walking distance, but going further again requires a car. Because the bus network is so badly integrated, anything more than a simple radial journey (to and from the city centre) becomes a nightmare.

The contrast with Kontula is dramatic. In the centre of the area is a large shopping centre and a metro station; in the focus groups all participants considered this within walking distance. Certainly Kontula is hardly a tourist attraction: the metro station is just about clean, but bare and functional with some graffiti. Nonetheless, the area as a whole is well served by buses and in addition there is a so-called ‘service-line’ or smaller internal bus-line. Everybody lives within 500 m of a bus-stop, most within 250 m. One metro stop away is Itäkeskus, a huge shopping come service centre, with all the main department stores and also all public facilities. And on the metro Kontula is only 17 minutes away from Helsinki city centre.

In Kontula both the survey responses and the focus group discussions show that for most people car ownership was an optional extra, a luxury and not a necessity, unlike in Jobstown. Thus, 50 per cent of those employed travelled to work by car, but nearly all of them considered that alternatives were available; Figure 3 shows that Kontula had one of the lowest levels of travel to work car dependency in the study. Unlike in Jobstown, owning a car did not usually enable people to access qualitatively better employment opportunities. Equally, with basic shopping, entertainment and healthcare within reach, not having a car did not pose problems for daily life. In the survey three quarters of those who did not own a car disagreed with the statement ‘I would buy a car if only I could afford one’. Indeed, not having a car is often seen as a way of avoiding unnecessary expenditure.

In Jobstown and Kontula cars are used in different ways. In Kontula many of the cars visible on the streets are ‘summer cars’, which are only used in the summer since maintaining them through the cold and wet of the winter would be too expensive. When non-car owners in Kontula were asked what they would gain from having a car, virtually no one mentioned access to employment. The forgone benefits of car ownership were travelling to the country and other forms of leisure. Similarly, when car owners in Kontula were asked about the problems they would face if they did not have a car, they hardly mentioned employment but over two-thirds mentioned going to the country. In line with this, Figure 3 highlights the low overall level of car usage for non-work journeys in Kontula, but shows that for these journeys car dependency is relatively high.

Urban planners and architects have focused on the detrimental effects of automobility for civic life. While this is a powerful argument, empirical research suggests that the issue is not straightforward, largely because of the issue of safety. The centro storico of Bologna has become an international showpiece for car restraint. Yet compared to the inhabitants of all 11 other areas in our study, the inhabitants of inner city Bologna were most likely to feel unsafe in public spaces. The study did reveal some evidence that public transport itself is seen by transport users as a public space, but again this is undermined by concerns for safety and by the threat of vandalism. Nonetheless, the extensive public transport system of Helsinki does contribute to a fuller use of public space in suburban Helsinki, while in Dublin suburban space becomes dead space.

The contribution of public transport to social inclusion is much clearer. Here what matters is not simply individual journeys, but the extent to which public transport forms a system that links possible activity sites. Of the case study cities, Bologna, Dublin and Helsinki all had areas of working class public housing which were outside the city core and which were segregated in terms of function. In other words, all three areas broke the basic precepts of the new urbanism. However, in Kontula, even if they were on low incomes and/or unemployed, those without a car remained able to participate in the life of the city. By contrast, in Jobstown those without a car were isolated: getting a job or even doing the shopping became major and sometimes insuperable logistical problems. A public transport system physically ties a city together and enables all inhabitants of the city to move around the city. Such transport welfare enables inhabitants to be citizens: transport links are the sinews of urban citizenship.

1. There is too much traffic going through this area.

2. In this area cars are a danger to children playing outside.

3. I personally feel safe from any physical attack (e.g., mugging) while walking around this area.

4. I personally would not feel safe walking around this area in the evening.

5. In this actual area there is too much traffic noise.

6. In this actual area cars make a lot of pollution.

7. In this area roads are too dangerous for children to walk to school by themselves.

8. If I walk around this area I nearly always meet someone I know.

James Wickham is Jean Monnet Professor of European Labour Market Studies in the Department of Sociology, Trinity College Dublin, where he directs the Employment Research Centre. He has published on economic development and the labour market in Ireland and also researched urban transport and sustainability.