1 Timo Toivonen ja Taru Virtanen (toim.) Ajasta paikkaan: taloussosiologisia tulkintoja - From Time to Place: Interpretations from Sociology Sarja Keskustelua ja raportteja/ Series Discussion and Working Papers 3:2004 ISBN: 951-564-171-3 (pdf) 951-564-170-5(sid.) ISSN: 1459-7632 (pdf) 0357-4687 (sid.) 1 ESIPUHE Turun kauppakorkeakoulun taloussosiologien perinteinen kesäseminaari pidettiin Nauvossa, Seilin saarella Turun yliopiston Saaristomeren tutkimusasemalla 6.–7. elokuuta 2003. Osallistujina oli paitsi Turun kauppakorkeakoulun opettajia ja jatko-opiskelijoita niin myös taloussosiologeja tai taloussosiologiasta kiinnostuneita muista yliopistoista, ja kaukaisin vieras oli Brasiliasta asti. Seminaarilla ei ollut mitään selvää teemaa, vaan esitetyt paperit liittyivät hyvinkin erilaisiin aihepiireihin. Aihepiirien ja lähestymistapojen monimuotoisuus tällaisissa seminaareissa ei ole mitenkään haitaksi. Asia on pikemminkin päinvastoin, sillä näin ikään kuin pakotettiin osallistujat tutustumaan ja ottamaan kantaa normaalien aihepiiriensä ja lähestymistapojensa ulkopuolelle meneviin taloussosiologian osa-alueisiin. Tärkeintä onkin tällaisissa seminaareissa esitysten laadukkuus. Sen verran yhtenäistä teemaa kuitenkin oli, että tämän julkaisun otsikoksi on voitu valita "Ajasta paikkaan - taloussosiologisia tulkintoja". Paikkoihin viittaavat esimerkiksi julkaisun turismiin liittyvät artikkelit eli Antti Honkasen, Pekka Mustosen ja Taru Virtasen kirjoitukset. Paikkaan liittyy myös luonnollisesti Sueila Pedrozon nuorten kulutusta Rio de Janeirossa koskeva artikkeli. Selkeästi aikaan taas liittyvät ajankäytön muutoksista kertovat Pekka Räsäsen ja Timo Toivosen artikkelit, mutta myös yhtä nykyajan tärkeää kulutuksen trendiä eli vihreää kulutusta pohtiva Leena Haanpään kirjoitus sekä käsitteellis-metodologisesta näkökulmasta muutoksen tutkimusta käsittelevä Juho Härkösen artikkeli. Tärkeitä sosiaalisen toiminnan selittämiseen liittyviä kysymyksenasetteluja eli rationaalista toimintaa ja institutionalismia käsittelevät Jani Erolan ja Antti Gronowin artikkelit. Seminaarissa käytiin jokaisesta paperista vilkasta keskustelua, joka ulottui myös lounas- ja päivällistauoille ja siitä edelleen aina saunaan ja virvokkeiden nauttimiseen asti. Tämä oli artikkelien ensimmäinen ja suullinen arviointiprosessi. Toinen ja kirjallinen referee-prosessi oli ristikkäisarviointi siten, että allekirjoittaneen lisäksi jokainen artikkelin kirjoittaja kommentoi sähköpostitse jonkun toisen kirjoittajan artikkelia. Tämä toimikin hyvin, napakkaa sananvaihtoa syntyi ja artikkelit paranivat usein vielä huomattavasti. 2 Tämän julkaisun toimittajista Timo Toivonen on vastannut lähinnä sisällöllisistä kysymyksistä, kun taas toisena toimittajana ollut Taru Virtanen on vastannut lähinnä toimittamisen käytännöllisestä puolesta. Asia on vaativa varsinkin, kun kirjoittajia on useita kuten tässä tapauksessa. Turussa uudenvuoden aattona 2003 Timo Toivonen 3 SISÄLLYSLUETTELO 1 JANI EROLA: SUBSIDIES DEPENDENCE AND THE WORKING POOR – RATIONALIZATION OF ACTION AS (DIS)INCENTIVE TO WORK* .................................................................................................... 1 Abstract ..................................................................................................... 1 1.1 Introduction ...................................................................................... 2 1.2 The explanations of incentive effects............................................... 4 1.3 Bridging the hypotheses together..................................................... 6 1.4 How to measure rationality? ............................................................ 8 I The description of the used variables.................................................................9 II The test of dimensionality ..............................................................................12 1.5 The analysis of the hypotheses of (dis)incentive effects................ 16 I Simple utilitarian explanation ..........................................................................17 II Risk-awareness explanation............................................................................19 III Cultural explanations: subsidy dependence...................................................20 IV Cultural explanations: macho-effects............................................................20 1.6 Best explanation? ........................................................................... 22 References ............................................................................................... 25 2 ANTTI GRONOW: INSTITUTIONALISMI TALOUSTIETEESSÄ JA TALOUSSOSIOLOGIASSA ................................................................. 29 Tiivistelmä .............................................................................................. 29 2.1 Instituutioiden uusi tuleminen........................................................ 29 2.2 Keskustelu institutionalismeista..................................................... 31 2.2.1 Taloustieteen uusi institutionalismi..........................................................31 2.2.2 Taloustieteen vanha institutionalismi.......................................................35 2.2.3 Taloussosiologian vanha institutionalismi ...............................................39 2.2.4 Taloussosiologian uusi institutionalismi ..................................................41 2.2.5 Organisaatiotutkimuksen institutionalismit..............................................42 2.3 Instituutioiden pylväät.................................................................... 44 2.4 Instituutioiden taustaoletukset ja integroidun instituutiomallin mahdollisuus................................................................................... 49 Kirjallisuus .............................................................................................. 51 4 3 LEENA HAANPÄÄ: PERSPECTIVES ON CONSUMPTION AND ENVIRONMENT ................ 55 Abstract ................................................................................................... 55 3.1 Introduction .................................................................................... 55 3.2 Green consumption, consumers and the environment: definitions 57 3.3 Green consumption and social change ........................................... 60 3.4 Green consumption and micro-level consumers ............................ 62 3.4.1 Future perspectives on consumer studies within ecological modernisation theory ............................................................................... 66 References ............................................................................................... 67 4 ANTTI HONKANEN: MATKAILUN KAUSILUONTEISUUS ........................................................ 71 Tiivistelmä............................................................................................... 71 4.1 Kausiluonteisuus matkailun ongelmana......................................... 71 4.2 Sosiaalisten käytäntöjen merkitys .................................................. 73 4.3 Aineisto........................................................................................... 74 4.4 Kausiluonteisuuden mittaaminen ................................................... 75 4.5 Kausiluonteisuus Euroopan unionissa vuosina 1985 ja 1997 ........ 79 4.6 Kausiluonteisuuden muutos ........................................................... 81 Kirjallisuus .............................................................................................. 83 5 JUHO HÄRKÖNEN: AIKA, SOSIAALINEN KAUSAATIO JA RAKENTEELLISET VAIKUTUKSET VERTAILEVASSA TUTKIMUKSESSA ................ 89 Tiivistelmä............................................................................................... 89 5.1 Johdanto.......................................................................................... 89 5.2 Aika ja kausaaliset mekanismit ...................................................... 90 5.3 Metodologiset ratkaisut ja aika vertailevassa makrososiologiassa 94 5.4 Aika, ja siirtyminen makrosta mikroon.......................................... 96 5.5 Lopuksi ......................................................................................... 100 Kirjallisuus ............................................................................................ 102 5 6 PEKKA MUSTONEN: PERSPECTIVES ON POSTMODERN TOURISM AND AUTHENTICITY ................................................................................. 105 Abstract ................................................................................................. 105 6.1 Introduction .................................................................................. 105 6.2 Postmodern tourism and tourists .................................................. 106 6.3 Some alternative thoughts ............................................................ 110 6.4 Post-tourists searching for authenticity? ...................................... 113 6.5 The challenge of sustainability..................................................... 115 6.6 Discussion .................................................................................... 117 References ............................................................................................. 119 7 SUEILA PEDROZO: INFORMATION TECHNOLOGY AND CONSUMPTION HABITS AMONG YOUNG PEOPLE FROM RIO DE JANEIRO .................... 123 Abstract ................................................................................................. 123 7.1 Introduction .................................................................................. 123 7.2 Theoretical Background ............................................................... 126 7.3 The Survey ................................................................................... 127 7.4 Work, Money and Consumption – Young People and Families.. 128 7.4.1 Respondent as Consumer/Individual Consumption ...............................133 7.4.2 Mobile Phone and Consumption Styles .................................................133 7.5 Young People, Information Technology and Consumption ........ 138 7.6 Consumption, Security and Environmental Consciousness......... 142 7.7 Conclusions .................................................................................. 144 References ............................................................................................. 146 8 PEKKA RÄSÄNEN: THE CHANGING CHARACTERISTICS OF COMPUTER USE IN FINLAND 1987-2000........................................................................... 148 Abstract ................................................................................................. 148 8.1 Introduction .................................................................................. 148 8.2 Approaching computer use from a cultural perspective .............. 150 8.3 Possession of home computers in the late 1980s and 1990s........ 151 8.4 Time use on computing in the late 1980s and 1990s ................... 154 8.5 Has computing become a ubiquitous activity?............................. 157 References ............................................................................................. 159 6 9 TIMO TOIVONEN: POLARIZATION IN TIME USE? SOCIO-DEMOGRAPHIC BACKGROUND OF THE INCREASE IN TELEVISION WATCHING IN FINLAND ................................................................. 161 Abstract ................................................................................................. 161 9.1 Introduction .................................................................................. 161 9.2 Theoretical background................................................................ 162 9.3 Data and variables ........................................................................ 164 9.4 Results .......................................................................................... 166 9.5 Summary and discussion .............................................................. 175 References ............................................................................................. 176 10 TARU VIRTANEN: ANALYZING CULTURAL TOURISM IN EU: YOUNG ADULTS’ CULTURAL ACTIVITY AND CONSUMPTION WHILE ON HOLIDAY............................................................................................. 179 Abstract ................................................................................................. 179 10.1 Introduction .................................................................................. 180 10.2 Cultural Tourist Meets Cultural Capital in Social Context .......... 180 10.2.1 Good Taste and High Culture – The Formation of Cultural Capital ..... 182 10.2.2 Research Questions and Data ................................................................ 183 10.3 Conceptualization of Cultural Capital in Tourism – Variables Used in the Analysis..................................................................... 184 10.3.1 Cultural Consumption Variables ........................................................... 184 10.3.2 Demographic Variables ......................................................................... 186 10.4 Methodology ................................................................................ 187 10.5 Empirical Findings on Young Adults’ Cultural Tourism............. 187 10.5.1 Consumption While on Holiday ............................................................ 189 10.5.2 Sand or Sidewalks? Choosing between Sea and City Destination ........ 193 10.6 Conclusions .................................................................................. 196 10.6.1 Peculiar young travelers from the Nordic countries .............................. 196 References ............................................................................................. 198 Appendix ............................................................................................... 201 KIRJOITTAJIEN YHTEYSTIEDOT – THE CONTACT INFORMATION OF THE AUTHORS…………………………………………………210 1 1 SUBSIDIES DEPENDENCE AND THE WORKING POOR – RATIONALIZATION OF ACTION AS (DIS)INCENTIVE TO WORK* Jani Erola Abstract Some people choose to work even if welfare benefits would guarantee them higher income than the low-income work. What explains this? Are the ones working with low income valuing working so much more positively than being without work that they choose to work even if they were better off as unemployed? Or do they just pay more attention to other risks connected to being outside normal employment, being thus more risk-aware than other persons? If compared to low-income workers, are the persons in the households without work or capital income more clearly instrumentally oriented, choosing not to work because of sufficient income even without spending their time in working? These assumptions are studied with Finnish national-level survey data “Finland 1999”. In order to test the hypotheses, the dimensions of rationality of action are derived empirically from a set of attitude variables. The low- income workers are compared to persons in families without work or capital income as their main source of income, as well as to rest of population outside these two groups. Three groups of hypotheses are considered: the utility related (simple utilitarian and risk-awareness), cultural (subsidies dependency and macho- effects) and structural explanations. The analysis conducted with log-linear models supports best the structural explanations. KEYWORDS: rationality, utility, risk-awareness, working poor, subsidies dependency, macho-effect * An earlier version of this article has also been presented at the 6th Conference of the European Sociological Association (ESA), in Murcia, Spain, 23-26th September, 2003, in the Economic Sociology Network Stream 5b ‘Economic Sociology and Economic Institutions’. 2 1.1 Introduction Through the history of the welfare state a central part of the criticism targeted at it has been the claims about the disincentive effects of the subsidy system on people's willingness to work. According to this interpretation of the welfare state, the type of society has itself created part of its problems by causing too strong “dependence” on the welfare subsidies. The welfare system itself is assumed to create strong disincentives to work especially in certain low income groups. Even such authorities as OECD has suggested that high rate of unemployment may be caused by too high rate unemployment benefits (see Drøpping et al 1999). In recent years some empirical support has also been found for the phenomenon. Atkinson et al studied the topic comparatively in the late 1980s, finding some support for the assumptions, although the institutional differences between the studied countries (Denmark, Sweden, West-Germany & UK) were significant (Atkinson – Mogensen 1990). Recent empirical results from Denmark (Pedersen – Smith 2002) suggest that the level of unemployment benefits effects on the tendency of being unemployed, although the majority of population in this kind of situation still chooses to work. As a general case, the harming functions of unemployment are considered to be so great that they are very hard to be rationalized by economic calculation (see Whelan – McGinnity 2000). Further, research results (for example McGinnity 2002) suggest that the problem is to be witnessed especially as a higher likelihood of both adults to be unemployed. This is found to be at least partially explained by the means-tested benefit systems. However, similar results have also been found in Finnish study (Virmasalo 2002), although Finnish beneficiary system is mainly based on individual income-tested benefits. While the “freeriding” problems connected to the benefit systems are underlined in research, there has not been so much interest towards the other side of the coin, namely why some persons nevertheless choose to work, even if they would do economically better by being jobless and living on subsidies. Table 1 from dataset “Finland 1999” exemplifies the phenomenon. The table presents the mean of income for the persons in the families where all adults are unemployed (group 1), for the persons in the families completely depending on subsidies and being without work or capital income, but not necessarily unemployed1 (group 2), for the persons who are currently working (group 3), and for persons in the families depending on work or capital income 1 For example one adult is unemployed, while another adult in household is a pensioner or on a maternity leave. 3 and themselves currently working but being in the lowest income quintile (group 4). The income is presented in euros (converted from original FIM) and standardised with (older) OECD equivalence scale. The persons over 62 years old, the old age pensioners2, the persons living with their parents and students are excluded from the analysis. Also in order to avoid the problems of distinction to be made between dual-earner vs. single-earner families, the person included in the groups of dependent of work income are working themselves. The table shows that persons working but being in lowest income quintile actually have lower mean income than persons currently unemployed and depending on welfare subsidies. There is also a major difference between the income levels of the two jobless groups depending on welfare subsidies, suggesting that it may be valuable to consider other ways of exiting work force also than just unemployment. What explains that people end up in different groups presented in Table 1? It can be argued that if the incentive and disincentive effects play a role they are often assumed to, one should be able to explain the differentiation of these groups with these effects. Are the ones working with low income, the ”working poor”, more clearly value rational, valuing working so much more positively than being without it that they choose to work even if they were Table 1. Mean of income in euros and it's standard deviation in Finland 1999 data for (1) persons in families where all adults are unemployed, for (2) persons in families completely dependent on subsidy income, but in which not all adults are unemployed, (3) persons, who are currently working, and (4) persons who are currently working but remain in lowest income quintile (the “working poor”). Income standardised with (older) OECD equivalence scale. Persons over 62 years old, old age pensioners, students and persons living with their parents are excluded. N Mean Std.Err. 1) All adults in family are unemployed 65 547.57 70.49 2) subsidies, both not unemployed 97 756.60 59.04 3) working 894 1145.94 18.40 4) working, income in work or capital, lowest quintile 62 457.39 20.55 2 The usual age limit for old age pension in Finland is 63 years in public sector and 65 in private sector. However, there are also some ways to exit to old age pension also younger than that, although this usually means that the pension will be reduced. 4 better off without? Or do they just pay more attention to other risks connected to being outside normal employment, being thus more risk aware than other persons? Are the persons in households without work or capital income more clearly instrumentally oriented, concerning about proper means in relation to wanted economic outcomes, and choosing not to work because of sufficient income even without spending their time in working? In this paper the groups of Table 1 are studied as “incentive groups” and analyzed from the point of view of rational behaviour. The structure of this paper is following. First the three types of explanations of disincentive and incentive effects are outlined. Then a rational theory covering all of these explanations is proposed. After that empirical content for the theory is acquired from the attitude variables of Finnish national level dataset “Finland 1999” by using multidimensional scaling. Then the different explanations are tested using the proposed theory. 1.2 The explanations of incentive effects There are at least three groups of explanations attached to the incentive effects: the utility related explanations (including simple and risk related utilitarian explanations), the cultural explanations (under- or macho culture related explanations) and the structural explanations. According to the simple utilitarian economic interpretation it would be assumed that the beneficiary system is too generous in welfare states and does not encourage jobless persons to work, because they already have efficiently high income level because of welfare beneficiary system. This is approximately the point of view that above was connected to OECD’s guidelines (see Drøpping et al 1999). According to it the persons, who nevertheless “choose” to work with low salary income, are assumed to be less economically rational than other groups. If there is a strong structural explanation of unemployment (i.e. lack of jobs) – like it has been the case in Finland from the early nineties, if the ge- neral unemployment rate is considered – becoming unemployment cannot be easily considered as voluntary process3. In this kind of situation, if the utilitarian explanation applies, the persons who are employed after being already unemployed are less likely to be economically rational than those who remain unemployed. This is because the persons applying instrumental rationality (understood as selection of proper means to acquire wanted 3 In general it often is the case that the process that leads persons to the two jobless incentive groups in Table 1 is probably most often ”involuntary”. 5 outcomes) in the pursue for economic utilities would prefer the straight economic awards related to unemployment over the one’s gained from a low- paid job, especially if the social costs of unemployment are relatively low because of many others are also unemployed. If the utilitarian explanation applies, the working poor should be less likely to be instrumentally rational than the persons of both jobless incentive groups. Also, as the income level of the families in which all adults are not unemployed but nonetheless depending on subsidies seems to be higher than in the families in which all adults are unemployed, the persons in the previous group should also be more likely to be instrumentally rational than persons in the latter group. Another type of utility-related explanations would be the risk-avoidance principle. According to it working with low income is preferred to being unemployed because of the high risks associated with unemployment. This explanation is in principle supported by the results of Whelan and McGinnity (2000). If the risk-avoidance principle would be a correct hypothesis as an explanation of what distinguishes the groups of Table 1, the persons with low income from work should be more clearly risk-aware when compared to groups outside of the workforce. At least this phenomenon should be observable in the group of the persons who previously have been unemployed, but are currently employed. The second branch of explanations is the cultural explanations. The cultural explanations can have various forms, for example underlining a certain kind of underculture of unemployment, according to which being unemployed is normatively accepted as one life-course choice, or the “macho-effect” explanation, according to which women are “forced” to leave their jobs because it is embarrassing for males in these families that women are breadwinners. What combines these explanations is that they underline the existence of strong normative or value base that hinders people getting a job. For example, if the underculture explanation would be correct, this should be observed as a stronger likelihood to value-oriented action in both groups of persons in jobless families, at least when compared to the working poor. Similarly, if macho-effect explanation holds, value-based action should be more likely for men especially in the jobless households. Finally structural explanations can also be considered. These explanations can be differentiated from the previously described types by the fact that while the previous underline the voluntary aspect of incentive effects, the structural type underlines deterministically behaviourist explanation of rational behaviour in relation to incentive effects. In the core of the latter type of explanations is the assumption that the differences in rationality of persons do not vary too much on the different levels of structural variables. For example, if there are no differences in rationale of action according to incentive group 6 or unemployment experiences, this would suggest the labour market related explanation to be correct, especially if the “spouse-effect” is controlled. Also the group of homogamy-effects may be considered as an example of “structural” explanations – people tend to marry to other people close to their own status, which may explain the stronger likelihood of both adults in household to be unemployed than it would be otherwise (See De Graaf – Ultee 2000; Härkönen 2003). As an extreme case, the structural explanation can be seen as parallel to any rationality related explanations of incentive effects. For example, if utility related explanations apply, this means that structural reasoning is not sufficient condition as explanation of people ending up in incentive groups of Table 1. However, structural explanation does not necessarily assume that there are no differences in rationalities between different groups. The structural explanation can be considered as valid also if there are differences in rationality of groups, but these differences fail to provide a cause for a person ending up in a certain incentive group. This would mean that the question is merely about rationalization of behaviour that is possible in certain environmental context rather than about rational voluntary choice made more or less freely. 1.3 Bridging the hypotheses together All the explanations can be tested against each other by applying a general theory of rational behaviour that is able to include all the types of action that were assumed in the different theories of the incentive effects. The idea is to use the types of rational action as “bridge hypotheses” combining the lower level theoretical statement to the more general level “core hypothesis”. In this way even very limited versions of rational action, such as some of the strictest versions of rational choice theories (RCT), are assumed to reach the four criteria of Hempel’s and Oppenhaim’s deductive-nomological explanation: the explanandum logically derived from explanans, the explanans containing at least one law, the explanans with empirical content and the explanans that is also true. (Kelle – Lüdemann 1998.) Taking into account the recent enthusiasm and the critique towards the sociological rational action theory (see Goldthorpe 1996; 1997; 1998; Edling 2000; Boudon 2001b; Ultee 2001; see Blossfeld & Prein 1998), what kind of a theory of rational action is needed, then? The theory should have the ability to include instrumentality and value orientation, because these types are required by the utilitarian and the cultural assumptions. Also, the theory should include risk-awareness as a possible type action, as the risk-awareness hypothesis 7 assumes. Thereby it is required that rationality is allowed to have variation without being considered as irrational behaviour. If the pattern of variation can also be verified empirically (as here by constructing the scales of rationality), the additional bridge hypotheses outside the theory are neither needed nor necessary. The general law-like assumption inside the theory is simply that human action in general tends to be meaningfully rational, in which empirical content – how this meaningfulness can be arranged as scales of rationality - is connected directly. What is gained is a simple explanation, although the theory allows more internal heterogeneity - and thereby more complexity - than for example utilitarian RCT. (See Hechter 1998.) What this kind of very general level rational action theory could do is that it can be used in verification of lower level theoretical statements that contain explicit statements of individual social behaviour and social mechanisms and structures, but which do not seem to have very obvious common grounds if considered separately, such as in the case of incentive effects. Max Weber’s classical theory of rationality of action contains two of the required types of rationality. In Weber’s text the first of them, instrumental rationality, is understood as action in which proper means are selected in relation to wanted outcomes of behaviour. Economic utilitarian rationality is assumed to require instrumentality, and can thus be considered to be special case of instrumental rationality. The second type, value-orientation, is understood as behaviour in which the action is guided by some important values. Values can also be used in order to put competing means in the order of importance. (Weber [1920]1978, 24–25, 63.) However, as Raymond Boudon has noted, Weber also acknowledged the possibility of action that would be guided purely by the axiological value of action itself, regardless the consequences. According to Boudon this is a type of rational action that usually has been missing from the followers of Weber. (Boudon 2001a; see also 1998.) What nevertheless seems to be missing from Weber’s theory is risk- awareness. This is quite an obvious shortage in Weber’s theory from today’s point of view, as the decisions made in risky situations are usually included even in much more limited versions of rational behaviour. This is where Ulrich Beck’s idea of risk rationality comes into picture. According to Beck, risk rationality should be considered as an ideal type of rationality that will become more usual in late modernity. Instrumental and value rationality are absorbed into a single ideal type of risk rationality, as the super-dangers, such as nuclear disasters, diseases and things like gene-manipulation, become more commonplace. When different expert groups fight over the right to define risks, truth itself becomes under dispute. (See Beck [1988]1990, 131–138.) 8 Although it is hard to prove whether it is true or not that risk-awareness has become more commonplace in recent decades, there is a more important finding that indeed makes risk rationality as its own ideal type. Beck points out that instrumental rationality needs a horizon of outcomes, but that with risks the horizon of future outcome of action is lost or at least becomes speculative (ibid., 135-136). Applying Weber’s terminology, this simply refers to an ideal type of action, where the outcome or consequences cannot be agreed on in similar manner that in instrumentality, simply because one does not know enough about all the possible consequences connected to be different means. Thereby, means loose their connection to ends or at least to knowledge about them. The means are selected as before, but actors are not bound to outcomes of them. (See Erola 2001.) However, if defined in this way, one does not need to assume a zeitgeist explanation in order to find risk rationality. It may be assumed that risk rational behaviour can be found in any life situations, where the obvious bounds to continuity are lost, such as in the case of social crises, like wars, economic depressions and natural disasters, or in the case of individual crises, like sicknesses, divorces or death of the family member. Risk rationality comes close to, for example, the idea of risk-aversion or risk-sensitiveness (see Tversky – Kahneman 1979; 1991; Kahneman – Lovallo 1993, Luhmann [1991]1993; and Goldthorpe 2000). A set of attitude questions are applied in order to get empirical content to the ideal types of rationality. Although many proponents of RCT argue in favour of using values and attitudes related measures as a means for validating the theory in use (Hechter – Kanazewa 1997; Opp 1999; Sen 1977; Hechter 1992; for the opposite arguments see Stigler & Becker 1977), there doesn’t seem to be too many studies actually applying such methods. Not even if for example Opp (1999) has noted that the selection of behaviour-related assumptions within theory should be primarily an empirical question, and that if the empirical means would be applied, the whole question about “thin” and “thick” versions of rational choice could be overcome. 1.4 How to measure rationality? In order to test the basic structure of rational behaviour, a special question pattern was included in the national level survey “Finland 1999”. The data set will be used through the analysis4. As there has been shown to be serious 4 The survey was collected in the spring and early summer of 1999. The original sample size was 4001, from which 61 percentages were covered in the final data. Persons under 18 or over 75 were not 9 problems in survey measurement of attitudes and values (see Fischhoff 1991), the usage of this approach needs to be described throughout. It was assumed that in order to study rationality in its whole range, action has to be considered in several occasions typical to individual life course. If not experienced by answerers themselves, the situations to be considered were selected in such way that most of people can still have opinions about them or position themselves in such interaction situation. I The description of the used variables Four hypothetical interaction situations were to be consider: making decision about taking a big loan (in order to buy a car or a house), making an important decision concerning family (such as moving to another location), making an important decision in work, and making an important decision concerning couple relationship. The questions were designed to be applicable to various decisions made in given situation, in order to make it easier for answerers to position themselves into it. There were also similar questions concerning buying things necessary for everyday life and choosing the way of payment in shop. These questions are, however, excluded from the current analysis, as it turned out that these tasks were conducted routinely and should not be considered as rational decision making in the similar manner that other variables can, because they were much more clearly everyday phenomena. It has been found elsewhere also that standardised questionnaires apply poorly to “routine” decision making in everyday life (Friedrichs – Opp 2002)5. Five types of statements were assumed to be able to cover the three assumed aspects of rationality (value orientation, instrumentality and risk- awareness). The answerers were asked if they agreed or disagreed with statements presented. The level of agreement was estimated with likert-scale: 1 – agrees strongly, 2 – agrees, 3 – does not agree or disagree, 4 – disagrees, 5 – disagrees strongly. Because there was not option “don't know”, in such cases included in the sample. The survey can be considered to be fairly successful; until today it has produced over twenty reviewed publications and monographs concerning wide range of topics of everyday life and consumption in Finland. From 2004 onwards the dataset will be freely available for the use of social scientists from Finnish Social Science Data Archive, see www.fsd.uta.fi. 5 Friedrichs – Opp (2002) found out that a standardised method of inquiring about a decision making system in everyday life worked out poorly, and that only after applying non-standardised method the test was successful. (ibid. 2000.) However, the obvious side-effect from this is that the results had only very limited level of generality; the study was conducted only with students. Also, it may be argued that the every-day behaviour that are defined simply as ‘behaviour that individuals engage regularly’ is still too restricting to various definitions of rational behaviour, because also the important ‘big’ decisions should be considered, too, having considerably effects on relevancy of each type of rationality in question. 10 answerers chose option 3, or did not answer at all. In this way also the answerers that were reluctant to choose were made to make a decision. In order to make results more easily interpreted, the scales are reversed, in the way that higher value will always refer to stronger agreement with the statement in question. The first statement that was targeted at the evaluation of value-based choices in each hypothetical situation was formulated similarly in each context: “Before doing a decision I have to consider if x (taking loan, making the decision about family, work or relationship) is right or wrong according to some important principle.” The second statement was supposed to take risk-awareness into account. The inability to optimize outcomes of actions was the main target of these questions. The main problem in this case was that optimality should be presented as more general case than just to include economic optimum. In the case of work, the formulation was mildly: “Each decision has to be considered separately.” In the case of taking a big loan, an alternative way of consideration was to underline the insecurity that it was supposed to create in future. Thus, the statement in the case of taking loan was: “Taking a loan creates serous insecurity in the future.” Decisions concerning family or couple relationship obviously lacked the speculative aspect typical to closely economy-related decisions. In these cases, the formulation was simply: “The outcome of big decisions considering (life of family, couple relationship) is impossible to see in advance.” Third type of statements was supposed to refer whether the answerer believes that there is at least some who can have good and purposeful information enough about the decisions in interaction situation. The formulation is in the case of work: “I can trust what somebody else recommends as the best option when an important decision is to be made.” The recommendations were not assumed to have too strong influence in the case of taking a loan. Even if such instances could be found, the recommendations would quite likely be based on economic evidence, not roles or authorities of such actors. Thus this kind of question for taking loans was not included in the survey. In the case of decisions concerning family and couple relationship the differentiation between traditional and expert authority was wanted to be made clear. Thus, the statements were more heavily biased on expert knowledge, as Beck's risk society theory suggested. The statements were: 11 “When doing a decision considering (family, couple relationship) an advice of expert or other outsider is important.” The fourth type of questions was meant to refer to possibility of gaining knowledge at all beforehand about any interaction situation. In the case of family, work and couple relationship the formulation: “While doing the decision I can rely on the beforehand gained knowledge.” In the case of taking loan, the formulation was without “previously”, because it was felt that in order to be instrumentally informed actor in this case, information cannot be too old (like in the case of different interest rates etc...). Last type of variables was those that were meant to estimate lack of trust in future outcomes of actions. The variable was formulated similarly in all cases: ”I somehow take into account that my choose turns out to be failure in the end.” On could probably argue that this is simply something that wise person does – takes possible failure into account. However, this is exactly what makes risk rationality as its own type of action. It refers that neither the knowledge nor means are considered as efficient enough to gain the means. The described attitude variables are collected to Table 2 (for details, see Erola – Räsänen 2000). 12 Table 2. Description and abbreviations of attitude variables used in estimation of dimensionality of rationality Abb. Value orientation,… taking a big loan………...………………… VALULOAN - ‘’ - making an important decision concerning family……...……………………………… VALUFAMI - ‘’ - making an important decision concerning job……………………………………….... VALUJOB - ‘’ - making an important decision concerning couple relationship…..……………...…….. VALUCOUP Uncertainty ……….. related to taking a big loan…..…………… UNCELOAN - ‘’ - related to making an important decision concerning family………...………………. UNCEFAMI - ‘’ - related to making an important decision concerning job……………………………. UNCEJOB - ‘’ - related to making an important decision concerning couple relationship……...…… UNCECOUP Trust on ……............ experts, making an important decision concerning family ………...……………… EXPEFAMI - ‘’ - recommendation, making an important decision concerning job…………………... RECOJOB - ‘’ - experts, making an important decision concerning couple relationship………….. EXPECOUP Previous knowledge.. .can help to make the decision concerning taking a big loan…………………...……… KNOWLOAN - ‘’ - can help to make the decision concerning family……..……………………………… KNOWFAMI - ‘’ - can help to make the decision concerning job ………………………………………... KNOWJOB - ‘’ - can help to make the decision concerning couple relationship………….…..………... KNOWCOUP Prepare oneself to …failure of decision concerning taking a big loan.............................................................. FAILLOAN - ‘’ - failure of decision concerning family…… FAILFAMI - ‘’ - failure of decision concerning job……… FAILJOB - ‘’ - failure of decision concerning couple relationship……………………………….. FAILCOUP II The test of dimensionality The dimensions of rational behaviour are derived from the attitude variables with multidimensional scaling (MDS). An advantage of MDS over factor 13 analysis is that the dimensionality of data can be much more efficiently reduced, because it does not assume the linear connection between the analyzed variables. This is because MDS is based on distances between points, whereas factor analysis is based on vectors. (Schiffman et al 1981, 13; Davidson 1983, 213; Ludlow 1999, 962-975.) Whereas factor analysis is in principle limited to continuous variables, there are no limitations of the level of measurement of the analysed variables in MDS. Although MDS has been rather widely used in social psychology and research of education, usual factor analysis has been much more often used in analysis of attitudes and structures of choices. However, recent studies suggest that often MDS would be more advantageous method for analysis of appropriate scales (see Brazill – Grofman 2002; Maslovaty et al 2001). It has even been found that at least in some occasions factor analysis easily produces an additional “bogus” dimension to studied phenomenon in question (Brazill – Grofman 2002). Table 3 presents the results of test of dimensionality, the stress measures for each additional dimension assumed to be found in data, according to Normalized Raw Stress, Stress-1, Accounted Dispersion and Tucker’s coefficient of Congruence. The program PROXSCAL tries to minimize Normalized Raw Stress, whereas Stress-1 (the classic Kruskal’s stress) is presented, because it can be compared to Sturrock’s and Rocha’s (2000) MDS Stress evaluation table. The upper limits for fit according to dimensionality and number of items can be set with stress evaluation table, which means that researcher is not so much dependent of the “rules of thump” as it has previously suggested in MDS (see Kruskal – Wish 1978, 56). One should expect to find the optimal dimension from the point where Normalised Raw Stress drops significantly, if compared to lower level of dimensionality. Also, in a fitting model Stress-1 should be under the upper stress limit given in stress Table 3. Stress and Fit Measures for rationality variables of Finland 1999 data set in test of dimensionality with Multidimensional Scaling. 1-4 dimensions; Normalized Raw Stress, Kruskal’s Stress-1, Dispersion Accounted For, and Tucker’s Coefficient of Congruence. Dimensionality 1 2 3 4 Normalized Raw Stress 0.110 0.020 0.008 0.005 Stress-I 0.331 0.141 0.091 0.070 Dispersion Accounted For (D.A.F.) 0.890 0.980 0.992 0.995 Tucker's Coefficient of Congruence 0.944 0.990 0.996 0.998 14 evaluation table and Dispersion and Congruence should be expected to be over the limit of 0.90. (See Kruskal – Wish 1978, 48-60; Schiffman 1981, 10–13; Davidson 1983, 68–71; Sturrock – Rocha 2000.) Means - to - Values 1,50,0-1,5 R is ks - to - O ut co m es 1,0 0,0 -1,0 FAILCOUP FAILJOB FAILFAMI FAILLOAN KNOWCOUP KNOWJOBKNOWFAMI KNOWLOAN EXPECOUP RECOJOB EXPEFAMI UNCECOUP UNCEJOB UNCEFAMI UNCELOAN VALUCOUP VALUJOB VALUFAMI VALULOAN Figure 1. Rationality variables of Finland 1999 data set in two-dimensional space according to results of MDS. The preliminary assumption made in analysis is that the optimal dimensionality is between 1 and 4 dimensions. One-dimensional solution is clearly not supported by the analysis, if compared to two-dimensional. Normalised raw stress on two dimensions is only one fifth from one dimensional solution. Also D.A.F and Tucker’s coefficient are on the sufficiently high level with two dimensions. The stress of 0.141 is well under the upper limit of 0.269 for 19-item two-dimensional setup in Stress evaluation table (see Sturrock – Rocha 2000), and can thereby be accepted as sufficient. The positions of original variables on the dimensions of rationality that MDS produced are presented in Figure 2. The variables related to uncertainty of action are positioned on down centre, variables related to trust to 15 knowledge and experts on up centre, while value-related variables are positioned in right centre. The dimensions of the MDS can thereby be named as means-to-values -dimension (dimension 1) of rationality and risks (or failures)-to-outcomes dimension of rationality (dimension 2). The figure can thus be theoretically interpreted in such a way, that Weber's (1920[1978]) two ideal types of rationality, means and value rationality, are located at the opposite ends of dimension 1. In the footsteps of Weber, the assumption about economic rationality that RCT usually applies, is assumed to be part of the left-hand part of figure, although all the action in that area of figure cannot be considered to be economically rational. Weber did not distinguish the dimension 2 explicitly, although he limited his types of rationality to upper part of dimension, while concerning the outcome-orientation in respect of means-rational action only. This is often considered as problem of Weber’s rationality types (i.e. Parsons 1937). The risk rational behaviour (see Erola 2001; Beck 1991), or risk-sensitive decision making, pointed out by i.e. Tversky and Kahneman (1979; 1991), Kahneman and Lovallo (1993), Luhmann ([1991]1993) and Goldthorpe (2000), is assume to be located in the lower part of the figure. The variation on the different dimensions is independent of the other dimension. Both because of empirical as well as theoretical support, the two- dimensional setup is selected as an optimal starting point, when Finland 1999 –dataset is being analysed. If the Weberian interpretation is followed, the upper parts of Figure 2 are named as instrumental and value rationality. The lower left part of the figure is named as risk rationality. How about the last quarter of the dimensional table, then? It would mean that action is value-oriented, but that again, the outcomes of action are less important factors. As it was already mentioned above, Boudon (2001a, 93–94) has noted that basically already Weber considered also this ideal type when assumed that it could be possible to consider action which is only motivated by axiomatic beliefs, but that nevertheless is not itself a “routine” or “tradition”. According to the example Boudon, this area of the figure is named as axiological rationality. The results of MDS are used in order to create two scales that can be applied on the individual level of analysis. This can be done by using the variable level scale coordinates as weights of the individual level data. For example in the case of means-to-values -dimensions, scale position of a person is aggregated simply by multiplying an individual answer to each question, varying from 1 to 5, with matching scale coordinate and then by summing up all the weight-multiplied answers of one individual answerer form all nineteen rationality variables together. This way all the variables can be used in the aggregation of both scales, thus minimum loss of important 16 information is accomplished without fear of risking the scale validity6. Because log-linear analysis will be used as method of analysis, people are distinguished into two equally sized groups according to both variables; the means-to-values-rationality scale into the groups of means-rational and value rational and risks-to-outcomes-rationality scale into the groups of risk-rational and outcome-rational. 1.5 The analysis of the hypotheses of (dis)incentive effects In the following analysis the means-to-values dimension of rationality is indicated with A, risks-to-outcomes dimension with R. The equation for full & saturated log linear model with two dimensions of frequency table F would be AR ij R j A i AR ijF λλλθ +++=log where subscripts i=1,..,I indexes the categories of variable A and j=1,..,J indexes the categories of R. The equation is not identifiable without some restrictions. The usual way of making a log-linear model identifiable is to set the last category of variables to be equal to zero, i.e. as a reference class. (Agresti 1990; Powers – Xie 2000.) Observed cell frequencies are reproduced with the log-linear equation using some iterative estimation method, here by using the EM -algorithm (Dempster et al. 1977). Estimated cell counts are then compared to the observed ones and goodness of fit statistics are used to estimate if the model fits the data. The most widely used test, the likelihood ratio chi-square test (G2), is used in order to test the goodness of fit of the log- linear model. Tables also present the chi-square significance of models (p) and dissimilarity index (∆). Dissimilarity index can be understood as the smallest fraction of cases that need to be re-classified in order to make model fit correct (see Agresti 1990; Firth – Kuha 1999). It is considered as a standard overall model fit statistic for log-linear models. The smaller the figure is, the better the model fit also is. Analyses were conducted using the LEM program (Vermunt 1997). Equations for hierarchical log-linear models are presented in simplified form in Table 4 and in all the log-linear models from this on. This means that 6In a previous analysis of structure of rationality similar scales were constructed simply by aggregating the ordinal variables that according to main component analysis get high loadings in the same component (see Erola 2001). In that kind of solution there is however always the problem of how much information is actually missed with the excluded variables. Main component scores could have been used in scale computation, too. 17 lower-level interaction terms are suppressed from the equation if the higher- order interactions are presented. Table 4. Log-linear models for rationality variables (A,R), unemployment experiences (U) and incentive groups (I). G2 p d.f. ∆ 0. (independence) {A,R,U,I,,wei(UI)} 143.699 0.000 40 0.102 1a. (base) {A,R,UI,wei(UI)} 48.186 0.054 34 0.062 1b. {AR,UI,wei(UI)} 48.067 0.044 33 0.062 1c. {AI,R,UI,wei(UI)} 28.719 0.584 31 0.053 1d. {AI,R,UI,wei2(UI)} 17.001 0.980 31 0.071 2a. {A,RI,UI,wei(UI)} 45.585 0.044 31 0.062 2b. {A,RUI,wei(UI)} 34.281 0.061 23 0.049 2c. {A,R,UI,wei2(UI)} 27.488 0.778 34 0.111 2d. {A,RI,UI,wei2(UI)} 23.998 0.811 31 0.105 2e. {A,RUI,wei2(UI)} 20.036 0.640 23 0.094 3. {AIU,R,wei(UI)} 24.325 0.386 23 0.050 I Simple utilitarian explanation The analysis begins from the test of utility related assumptions. Log-linear models are presented in Table 4. Two rationality variables (A,R), a variable for the incentive groups (I) from Table 1 and a variable concerning unemployment experiences (U) are needed in order to test the validity of the explanation type. Variable U can have three levels, “no experiences on unemployment”, “has once”, or “has various”. Because it is logically impossible to all adults of household to be currently unemployed and at the same time to have no experiences from unemployment at all, this cell is excluded from the models. In the equations of models this has been notified with the abbreviation WEI(UI). Model 0 presents “independence-model” assuming no association whatsoever between the modelled variables. This would basically contradict both utility-related explanations, the simple utili- tarian as well as the risk-related. The model does not fit the data. Model 1a tests the “biologically necessary” interaction term between unemployment experiences (U) and incentive groups (I). The interaction is necessary because there is naturally higher possibility to have several experiences from unemployment if one is already currently unemployed, when 18 compared to other groups. The model does fit the data if only significance of G2 ratio is considered (chi2 improvement 95.513 with 6 degrees of freedom, if compared to basemodel), but cannot be considered as sufficient because of high level of dissimilarity index. Model 1b tests the basic assumption of two- way interaction between rationality variables, which should not normally apply to data because of application of MDS in construction of variables. As suspected, this interaction term does not improve the model fit (chi2 improvement 0.119 with 1 d.f.). Model 1c is crucial to the simple utilitarian hypothesis. It tests whether the likelihood of being in either in the group of value- or means-oriented is associated with the incentive group the person belongs into, despite the possible connection between the unemployment experiences and the incentive group. If there would not be significant interaction between values-to-means- rationality and incentive groups, utilitarian assumption should be abandoned. This is not the case. There are significant differences between the groups, as the improvement of chi-squared ratio when compared to base-model (19.467 with 3 d.f.) reveals. However, when the differences between the parameters of the interaction term AI in Table 5 are studied closer, it turns out that the connection is to the opposite direction that the simple utilitarian hypothesis assumes – those working are also most likely means rational, whereas those on subsidies or unemployed families are most likely value-oriented. This does not yet prove the incentive effect to be not supported within Finland 1999 -data. It is not known if the means-orientation would actually hinder the unemployed to become employed, or that value-orientation would promote working. In order to test this, Model 1d is restricted to only those who actually have experiences from unemployment. Model 1d tests the same model as 1c in these more restricted conditions, by excluding the cells with persons with no unemployment experiences from the analysis (models with WEI2(UI)-notation). Now the model fit as such is almost perfect, but the parameters for the interaction term and especially directions of them (see Table 6) do not change much. The likelihood of means orientation of the working poor is still at the higher level than those who are without work, although the difference is smaller. Thus, the simple utilitarian assumption does not find support from the data, although we could nevertheless argue that former unemployment somewhat reduces the likelihood of means-orientation of the persons in low-paid job. 19 Table 5. Parameters for interaction term AI in model 1c. Likelihood of means-orientation if compared to value-orientation Coeff. Std.err. Z-value Odds All unemp. -0.112 0.104 -1.079 0.894 Subs., not all unemp. -0.302 0.102 -2.953 0.740 Working poor 0.238 0.110 2.169 1.269 Working or capital 0.176 * * 1.192 * reference class Table 6. Parameters for interaction term AI in model 1d. Only those with unemployment experiences included in the model. Likelihood of means-orientation if compared to value-orientation Coeff. Std.err. Z-value Odds All unemp. -0.106 0.113 -0.931 0.900 Subs., not all unemp. -0.212 0.130 -1.635 0.809 Working poor 0.083 0.155 0.538 1.087 Working or capital 0.235 * * 1.265 * reference class II Risk-awareness explanation The other utility-related hypothesis, the risk-awareness hypothesis, is to be tested next. Model 2a (table 3) tests whether there are any differences on the second rationality angle, risks-to-outcomes-orientation of action, between the four incentive groups, as the empirical support for the hypothesis would require. The model fit does not improve from the base model significantly (chi2 improvement by 2.601 with 3 d.f.). Also differences on the higher three- way-interaction with rationality-, incentive and unemployment group variables are tested in model 2b. Neither the model with thee-way interaction term improves the model fit significantly if compared to fit of base model (chi2 improvement 13.903 with 11 d.f.). The differences do not seem to occur, although it was anticipated in the hypothesis that those who had experienced unemployment more than once and were currently in the group of the “working poor”, would also have been more clearly risk aware than others. It may again be worth to consider whether the differences are to occur only when persons with experiences from unemployment are considered. Model 2c presents a “new” base-model, which is the same as Model 1a, except those who have not experienced unemployment are excluded from analysis. The 20 more “limited” models can be separated again with notification “WEI2(UI)”. Model 2d is similar to Model 2a and Model 2e similar to 2b, except in both cases the never unemployed are excluded. Neither of the interaction Models 2d or 2e improves the model fit, if compared to new base-model 2c. Thus, also the other version of utility related explanations fails to gain support from the analysis. Whether persons are more likely to be risk-aware than outcome- oriented is independent on previous unemployment experiences as well as from which of the incentive groups the persons belong into. III Cultural explanations: subsidy dependence How about cultural explanations, then? Previous models already showed that value-orientation is much stronger in the subsidies-depending groups than in the groups that are currently working. This may very well suggest that there is such a phenomenon as a subsidies-depending underculture. However, this cultural explanation cannot be considered as “valid” cause unless the three- way interaction between value-orientation, previous unemployment experiences and either of subsidies-depending groups can be observed. This is because in order to cultural explanation to be valid, these values should prevent the employment of either of the unemployed groups, whereas lack of these values should explain the employment of those previously unemployed. Model 3 (Table 4) tests the three-way interaction model in question. The model improvement is not significant when compared to model 1c (chi2 improvement 4.394 with 8 d.f.). Thus, even if there is a “cultural” connection between incentive groups and value-orientation, this is not an explanation of the process of people ending up in different incentive groups according to unemployment experiences. IV Cultural explanations: macho-effects The macho-effects explanation can be considered next. In order to evaluate its effect, the means-to-values rationality variable should be tested with spouse- and incentive variables. The macho-effects explanation assumes that value- oriented men are overrepresented in households with two non-working adults. Thus there should be three-way interaction in the model associated especially with these groups. The variable “spouse-effect” (S) replaces the unemployment experiences-variable. It consists of three groups: singles, cohabitated men and cohabitated women. Table 7 shows log-linear models for these variables. 21 Model 4 is the independence model assuming no interaction between the tested variables. The results (the statistically significant G-squared, high dissimilarity index) show that this is a false assumption within current dataset. Model 5 jumps straight to three-way interaction model of means-to-values- rationality, spouse-effect and incentive groups. The fit statistic improves significantly, if compared to independence Model 4 (chi2 improvement 130.380 with 17 d.f.), which is of course not a surprise, as the model also includes all the two-way interactions also. However, this time three-way interaction is significant also when tested against the model with only lower level interactions. When the fit statistics are compared to simpler all two-way interactions model 6, it can be seen that latter is significantly worse (chi2 decrease 13.791 with 6 d.f.). Table 7. Log-linear models for means-to-values-rationality (A), spouse-effect (s) and incentive groups (I). G2 P d.f. ∆ 4. (independence) {R,A,S,I} 162.240 0.000 40 0.101 5. three-way {R,ASI} 31.860 0.103 23 0.042 6. all two-way A-I-S {R,AI,SI,SA} 45.650 0.025 29 0.050 The parameters for the interaction term ASI from Model 5 are shown in table 16. It can be seen that the observed interaction is not so straightforward that the macho-effect explanation assumed. Among the cohabitated men in families where all adults are unemployed the likelihood to value-orientation is higher than for women in same group. However, the likelihood of cohabitated men in all-unemployed families to be value-rational does not differ significantly from the group of singles. In the case of other group not working the situation is exactly the opposite: women are value-oriented with as high likelihood that men in unemployed families were, whereas men in this incentive groups seem to be more likely to means-efficient. Thereby it seems to be that the effect cannot be considered as a result of macho-effects, because the difference should be expected to be towards the same direction in both of the groups. At the same time the clearest separable group is the one of cohabitated women in all-unemployed families, who tend to be more clearly means- oriented than other groups. This result does not seem to fit to any proposed hypotheses. What explains that these women are more clearly means-oriented than other groups not working? The explanation needs to be searched outside the current research setup. In fact it turns out that women in this group are the 22 only group not having any children living at home. This is most likely to be the explanation for the lower likelihood to value-orientation in this group. Table 8. Parameters for three-way interaction term ASI (means-to-values rationality, spouse-effect and incentive groups) in model 5. Likelihood of value-orientation if compared to means-orientation Coeff. St.err Z-value Odds Cohab. men 0.295 0.251 1.175 1.343 Cohab. women -0.542 0.247 -2.190 0.582 All unempl. Single 0.247 * * 1.280 Cohab. men -0.100 0.154 -0.645 0.905 cohab. women 0.421 0.162 2.598 1.524 Subs., not all unempl. Single -0.322 * * 0.725 Cohab. men -0.181 0.166 -1.092 0.834 Cohab. women 0.186 0.174 1.066 1.204 Working poor Single -0.005 * * 0.995 Cohab. men -0.014 * * 0.986 Cohab. women -0.065 * * 0.937 Working & capital Single 0.079 * * 1.083 * reference class 1.6 Best explanation? All the tested explanations so far have remained unsupported by the analysis in their original forms. Utility related explanations, both simple utilitarian and risk-awareness, as well as the tested cultural explanations, under- and macho- culture, were not supported as the types of explanation would have assumed. Finally, also the structural effects need to be considered. The analysis of both utility related and both forms of cultural explanations showed that although there was a connection between the rationality variables and the incentive groups, it could not be shown that the type of rationality of behaviour itself would have hindered or promoted people's belonging into any particular group. The following may thus also be considered as concluding remark from current analysis: it would be more correct to assume the rationality variables as reflecting the possible structural positions, and also changing rather easily when also the structural position changes, than to assume the differences in rational behaviour to orchestrate in which incentive groups they belong into. 23 This could explain the differences between rationality of women and men in cohabitated, unemployed families; people choose best behavioural schema in which they adapt their attitudes; they are the “impartial spectators”, like already noted by Adam Smith ([1759]1974, 110–113; Boudon 2000, 210– 211). People are likely to change the pattern of their behaviour and the rules they make decisions for various different reasons, such as for having children, or at least adapt their motifs in a way that could make their new conditions more sensible for them (for example see van Deth – Scarbrough 1995). In a sense it seems to be that the differences that were observed between the different incentive groups can be better explained with rationalization of behaviour than concerning the cause of one’s belonging into these groups to be explained by rational and voluntary selection. Persons in unemployed families tend to be means-efficient if their life-situations give an opportunity for it. Then again, having children may turn out to be a constraint directing behaviour in general towards more clearly value-oriented type. This would suggest using a life-course related explanation with rationality related explanation. This also underlines the importance of concerning structural explanations of behaviour prior to voluntary reasons. However, the fact that findings support strong structural explanations may only be due to strong impact of structural changes associated with the recession of the early nineties in Finland, and be something that in “normal” conditions of welfare state would not be found. For example Atkinson and Mogensen (1993) note in the conclusions of their book on work incentives that the incentive effects may not be found in the conditions of economic downturn. Also, it is likely that while rationalization is an explanation of rationality in less advantaged positions, in more advantageous positions the rationality may indeed present a voluntary choice not so much determined by the structural limitations. However, persons in advantageous positions were all analysed as one whole groups of those working or depending work-income. The important group of the expected utilities should also be tested; it may be that for example the young persons choose to work with low income in the beginning of their careers because they expect the income to be improved in the future. In general, it may be argued that there are others than monetary utilities related to being in low-income work that were not tested here. The results might also change from the current if a proper longitudinal data could be applied, which could make it possible to analyse the rationality of behaviour before and after unemployment experiences, or if more detailed log- linear models applying design matrices would be applied. Nevertheless, the analysis conducted here suggests that one should be very conscious when applying utility or culture related explanations of the 24 likelihood of belonging into somehow less advantageous position in society, because they may easily turn out to be in some extent fallacious. 25 References Agresti, Alan (1990) Categorical data analysis. John Wiley & Sons: New York. Atkinson, A.B. – Mogensen, Gunnar Viby (1993) Welfare and Work Incentives. A North European Perspective. Clarendon Press: Oxford. Beck, Ulrich ([1988]1990) Riskiyhteiskunnan vastamyrkyt. Vastapaino: Tampere. Blossfeld, Hans-Peter & Prein, Gerald (eds.) (1998) Rational Choice Theory and Large-Scale Data Analysis. WestviewPress: Oxford. Boudon, Raymond (1998) Limitations of Rational Choice Theory. American Journal of Sociology, Vol. 104, No. 3, 817–828. Boudon, Raymond (2001a) The Origin of Values. Sociology and Philosophy of Beliefs. Transaction Publishers: New Brunswick. Boudon, Raymond (2001b) Which Rational Action Theory for Future Mainstream Sociology: Methodological Individualism or Rational Choice Theory? European Sociological Review, Vol. 17, No. 4, 451–457. Brazill, Timothy J. – Grofman, Bernard (2002) Factor analysis versus multi- dimensional scaling: binary choice roll-call voting and the US Supreme Court. Social Networks, Vol. 24, No. 3, 201–229. Davidson, Mark L. (1983) Multidimensional Scaling. John Wiley & Sons: New York. Dempster, A. P. – Laird N.M. – Rubin D. B. (1977) Maximin Likelihood from Incomplete Data via the EM Algorith. Journal of the Royal Statistical Society, Series B 39, 1–38. De Graaf, Paul M. – Ultee, Wout C. (2000) United in Employment, United in Unemployment? Employment and unemployment of couples in European Union in 1994. In: Welfare Regimes and the Experience of Unemployment in Europe, ed. by Duncan Gallie and Serge Paugam, 265–285. Oxford University Press: Oxford. Drøpping, Jon Anders –Hvinden, Bjørn – Vik, Kirsten (1999) Activation policies in the Nordic countries. In: Nordic Social Policy. Changing Welfare States, ed. by Mikko Kautto, Matti Heikkilä, Bjørn Hvinden, Staffan Marklund and Niels Ploug (eds.), 133– 158. Routledge: London. 26 Edling, Carl (2000) Rational choice theory and quantitative analysis. A comment on Goldthorpe's sociological alliance. European Sociological Review, Vol. 16, No. 1, 1–8. Erola, Jani (2001) Maksuhäiriöisyys ja luottamusrakenteiden kriisi sosiaalisessa toiminnassa. Sosiologia, Vol. 38, No. 3, 174–91. Erola, Jani. – Moisio, Pasi (2002) Jähmettyikö Suomi? Sosiaalinen liikkuvuus ja pitkäaikaistyöttömyys Suomessa 1970-1995. Sosiologia, Vol. 39, No. 3, 185–199. Erola, Jani – Räsänen, Pekka (2000) Suomi 1999: aineistonkeruu ja tutkimusseloste. In: Maksuhäiriöisyys laman jälkeisessä Suomessa, Jani Erola, 75-88. Sosiologista keskustelua B 36. Turku. Firth, David – Kuha, Jouni (1999) On the index of dissimilarity for lack of fit in log-linear models. Unpublished research paper, 28 July, 1999. Nuffield College, Oxford. Fischhoff, Baruch (1991) Value Elicitation. Is There Anything There? American Psychologist, Vol. 46, No. 8, 835-847. Friedrichs, Jürgen – Opp, Karl-Dieter (2002) Rational Behaviour in Everyday Situations. European Sociological Review, Vol. 18, No. 4, 401– 415. Goldthorpe, John (1996) The Quantitative Analysis of Large-Scale Data-sets and Rational Action Theory: For a Sociological Alliance. European Sociological Review, Vol. 12, No. 2, 109–126. Goldthorpe, John (1997) The integration of Sociological Research And Theory – Grounds for optimism at the end of the twentieth century. Rationality & Society, Vol. 9, No. 4, 405–465. Goldthorpe, John (1998) Rational Action Theory for Sociology. British Journal of Sociology, Vol. 49, No. 2, 167–192. Goldthorpe, John H. (2000) On Sociology. Numbers, Narratives, and the Integration of Research and Theory. Oxford University Press: Oxford. Hechter, Michael (1992) Should Values Be Written Out of the Social Scientist’s Lexicon? Sociological Theory, Vol. 10, No. 2, 215– 230. Hechter, Michael (1998) The Future of Rational Choice Theory and its Relationships to Quantitative Macro-sociological Research. In: Rational Choice Theory and Large-Scale Data Analysis, ed. by Hans-Peter Blossfeld and Gerald Prein, 281–290. WestviewPress: Oxford. Hechter, Michael – Kanazawa, Satoshi (1997) Sociological Rational Choice Theory. Annual Review of Sociology, Vol. 23, 191–214. 27 Härkönen, Juho (2003) Jobless couples in Europe: what do we know on the basis of previous research? In: Individuaalisuus, rationaalisuus ja sosiaalinen rakenne – Individuality, Rationality and Social Structure, ed. by Timo Toivonen, 47–67. Turun kauppakorkeakoulun julkaisuja, Keskusteluja ja raportteja / Series Discussion and Working Papers 3:2003. Turku School of Economics and Business Administration: Turku Kahneman, Daniel – Lovallo, Dan (1993) Timid Choices and Bold Forecasts: A Cognitive Perspective on Risk Taking. Management Science, Vol. 39, No. 1, 17–31. Kelle, Udo – Lüdemann, Christian (1998) Bridge Assumptions in Rational Choice Theory: Methodological Problems and Possible Solutions. In: Rational Choice Theory and Large-Scale Data Analysis, ed. by Hans-Peter Blossfeld and Gerald Prein, 112–125. WestviewPress: Oxford. Kruskal, Joseph B. – Wish, Myron (1978) Multidimensional Scaling. Sage Publications: Beverly Hills. Ludlow, Larry H. (1999) The Structure of The Job Responsibilities Scale: A Multimethod Analysis. Educational and Psychological Measurement, Vol. 59, No. 6, 962–975. Luhmann, Niklas ([1991]1993) Risk: A Sociological Theory. Walter de Gruyter: Berlin. Maslovaty, Nana – Marshall, Anne E. – Alkin, Marvin C. (2001) Teachers' Perceptions Structured Through Facet Theory: Smallest Space Analysis Versus Factor Analysis. Educational and Psychological Measurement, Vol. 61, No.1, 71–84. McGinnity, Frances (2002) The Labour-force Participation of the Wives of Unemployed Men: Comparing Britain and West Germany Using Longitudinal Data. European Sociological Review, Vol 18, No .3, 473–488. Opp, Karl-Dieter (1999) Contending Conceptions of the Theory of Rational Action. Journal of Theoretical Politics, Vol. 11, No. 2, 171–202. Parsons, Talcott (1937) Structure of Social Action. A Study in social Theory with Special Reference to a Group of Recent European Writers. The Free Press: New York. Pedersen, Peder J. – Smith, Nina (2002) Unemployment Traps: Do Financial Disincentives Matter? European Sociological Review,Vol. 18, No. 2, 271-288. Powers, D. A. – Xie, Y. (2000) Statistical Methods for Categorical Data Analysis. Academic Press: New York. 28 Sen, Amartya (1977) Rational Fools: A Critique of the Behavioral Foundations of Economic Theory. Philosophy & Public Affairs, Vol. 6, No. 4, 317–344. Schiffman, Susan S. – Reynolds, M. Lance – Young, Forrest W. (1981) Introduction to Multidimensional Scaling. Theory, Methods, and Applications. Academic Press: Orlando. Smith, Adam ([1759] 1974) The Theory of Moral Sentiments. Ed. by D. D. Raphael – A. L. Macfie. Clarendon Press: Oxford. Stigler, George J. – Becker, Gary S. (1977) De Gustibus Non Est Disputadum. The American Economic Review, Vol. 67, No. 2, 76–90. Sturrock, Kenneth – Rocha, Jorge (2000) A Multidimensional Scaling Stress Evaluation Table. Field Methods, Vol. 12, No. 1, 49–60. Tversky, Amos – Kahnemen, Daniel (1979) Prospect Theory: An Analysis of Decision under Risk. Econometrica, Vol. 47, No. 2, 263–292. Tversky, Amos – Kahneman, Daniel (1991) Loss Aversion in Riskless Choice: A Reference Dependent Model. Quarterly Journal of Economics, 1991, November, 1039-1061. Ultee, Wout (2001) RAT-QAD or PCP-AAD? European Sociological Review, Vol.17, No. 4, 451–467. Van Deth, Jan W. – Scarbrough, Elinor (1995) The Concept of Values. In: The Impact of Values, ed. by Jan W. van Deth and Wlinor Scarbrough, 21–47. Oxford University Press: Oxford. Vermunt, J. K. (1997) LEM: A general program for the analysis of categorical data. Tilburg: Tilburg University. Virmasalo, Ilkka (2002) Perhe, työttömyys ja lama. Jyväskylä studies in education, psychology and social research 204. Jyväskylän yliopisto: Jyväskylä. Weber, Max ([1920] 1978) Economy And Society. An Outline of Interpretative Sociology. University of California Press: Berkeley. Whelan, Christopher – McGinnity, Frances (2000) Unemployment and satisfaction: a European analysis. In: Welfare Regimes and the Experience of Unemployment, ed. by Duncan Gallie and Serge Paugam. Oxford University Press: Oxford. 29 2 INSTITUTIONALISMI TALOUSTIETEESSÄ JA TALOUSSOSIOLOGIASSA7 Antti Gronow Tiivistelmä Instituutiot ovat viime aikoina nousseet taloustieteen ja taloussosiologian keskeisiksi kysymyksiksi erilaisten uusien institutionalismien muodossa. Artikkelissa tarkastellaan näiden oppiaineiden uusien ja myös vanhojen institutionalististen lähestymistapojen erilaisia instituutio-käsityksiä. Taloustieteen ja taloussosiologian uudet institutionalismit edustavat regulatiivista instituutio-näkemystä, jossa instituutiot ovat ennen kaikkea yksilöiden toimintaa rajoittavia tekijöitä. Sosiologian vanhalle institutionalismille ja sosiologis-vaikutteiselle organisaatiotutkimukselle instituutiot ovat normativiisia instansseja, kun taas organisaatiotutkimuksen uusi institutionalismi pitää instituutioita kulttuuris-kognitiivisia merkityskehyksiä. Kaikista laajin instituution määritelmä löytyy kuitenkin taloustieteen vanhasta institutionalismista, joka sai vaikutteita pragmatistisesta filosofiasta. Tällaisen institutionalismin edustajan Thorstein Veblenin mukaan instituutiot pohjaavat ennen kaikkea habituaaliseen toimintaan. Artikkelissa selvitetään institutionalismien erilaisia taustaoletuksia ja pohditaan myös sellaisen mallin mahdollisuutta, jossa eri instituutio-näkemykset voitaisiin yhdistää. ASIASANAT: Institutionalismi, instituutiot, toimintateoria, pragmatismi 2.1 Instituutioiden uusi tuleminen Kulttuurintutkimuksen nousun myötä instituutiot ovat olleet osittaisessa teoreettisessa paitsiossa sosiologiassa. Silloinkin kun ne on otettu huomioon, on analyysissa yleensä jääty vain niin sanotulle kulttuuris-kognitiiviselle 7 Kiitän artikkelini aiempien versioiden kommentoinnista Risto Heiskalaa, Erkki Kilpistä ja Pekka Sulkusta. 30 tasolle, jolloin kaikesta merkityksestä tulee tietoa tai ”diskurssia”. Kulttuuris- kognitiivisella tasolla operoivat teoriatkin yleensä pyrkivät jotenkin huomioimaan reflektoimattomien tapojen olemassaolon, joilla ei ole tietoista tulkintaa, mutta seurauksena tästä on työskentely sellaisilla käsitteillä kuin ”hiljainen tieto”. Miten tällaisesta yhteiskuntateoreettisesti epätyydyttävästä tilanteesta päästään eteenpäin? Yksi keino on instituutioiden ottaminen eksplisiittisen teoretisoinnin keskiöön. Viime vuosina uudesta tai uusinstitutionalistisesta lähestymistavasta on keskusteltu ainakin taloustieteessä (Vihanto 2003), taloushistoriassa (Lamberg – Ojala 1997), taloussosiologiassa (Brinton – Nee 1998; Heiskala 2003a), sosiologisesti virittyneessä organisaatiotutkimuksessa (DiMaggio – Powell 1991; Scott 2001), valtio-opissa (Aranson 1998), antropologiassa (Ensminger 1998), kehitystaloustieteessä (Raumolin 1995) ja sosiaalipolitiikassa (Saari 2003). Tässä artikkelissa rajaan tarkasteluni kuitenkin koskemaan lähinnä taloustieteellistä ja taloussosiologista8 keskustelua, koska institutionalismien ja etenkin sosiologian kannalta keskeisimmät debatit on käyty juuri näillä oppialoilla. Osin keskustelussa on kyse taloustieteen ja sosiologian vanhasta kädenväännöstä, jota on käyty yhtä pitkään kuin oppiaineet ovat olleet olemassa. Voinee silti väittää, että keskustelussa ei ole kyse pelkästä vanhan toistosta. Uutta lienee ainakin sekä taloussosiologiasta että taloustieteestä yleisemminkin löytyvä yhteistyö- ja keskusteluhalukkuus. Osittain kyse on siitä, että taloustieteessä on herätty huomaamaan perinteisesti sosiologian alaan kuuluvien kysymysten, kuten instituutioiden, olemassaolo. Toisaalta taas sosiologit ovat esittäneet, että sosiologia pystyy selittämään taloustiedettä paremmin monia perinteisesti taloustieteen alaan luettuja ilmiöitä. Nämä selitysyritykset ovat kulkeneet uusi taloussosiologia -kattokäsitteen alla (ks. esim. Granovetter 1990). Eri oppialojen institutionalismit ovat hyvinkin erilaisia teoriaperinteitä, vaikka yhdistäviä seikkoja on toki löydettävissä. Taloustieteessä ja taloussosiologiassa institutionalismeja yhdistää ensinnäkin se, että ne ovat syntyneet talouden tarkastelun puitteisiin, ja toisekseen se, että niissä luonnollisestikin korostetaan instituutioiden roolia. Yhteinen lähtökohta niillä on myös siinä, että ne enemmän tai vähemmän kritisoivat uusklassista taloustiedettä - osa fundamentaalisemmin ja osa enemmän uusklassisen taloustieteen peruslähtökohdista yhä kiinni pitäen. Tyypillistä institutionalismi-keskusteluille on näiden seikkojen lisäksi vanhan ja uuden institutionalismin erottaminen toisistaan. 8 Luen artikkelissani taloussosiologiseen keskusteluun myös institutionaalisen organisaatiotutkimuksen siksi, että se nojaa sosiologiaan keskeisissä teoreettisissa oletuksissaan. 31 Institutionalismi-keskusteluista voidaan kaivaa niiden ”kova ydin” esiin tarkastelemalla sitä, millaista instituutio-käsitystä kukin teoriaperinne edustaa. Käytän institutionalismi-keskustelujen luokittelemiseen W. Richard Scottin (2001) lanseeraamaa jakoa instituutioiden regulatiivisiin, normatiivisiin ja kulttuuris-kognitiivisiin pylväisiin. Regulatiivisissa näkemyksissä korostetaan instituutioiden toimintaa rajoittavaa luonnetta, jolloin instituutiot nähdään eräänlaisina sosiaalisina sopimuksina ja sanktioina. Normatiivisten näkemysten mukaan instituutiot pysyvät pystyssä sanktiouhan sijaan ennemminkin yhteisöllisen paineen takia. Kulttuuris-kognitiiviset näkemykset taas ymmärtävät instituutiot kulttuurisina merkityskehyksinä, joiden ylläpitämisessä sanktioilla ei ole juurikaan roolia. Kyse on yhteisistä ja itsestäänselvinä pidetyistä käsityksistä ja merkityskehyksistä, jotka konstituoivat sosiaalisen todellisuuden. Suomessa Scottin jaottelua on esitellyt ja tietyiltä osin myös täydentänyt Risto Heiskala (2003a). Heiskala lisää Scottin jaotteluun neljännen pylvään, jota hän nimittää pragmatistiseksi institutionalismiksi. Sekä Scott että Heiskala tyytyvät kuitenkin käytännössä vain olettamaan jaottelun toimivuuden eri institutionalismien instituutio-käsitysten ryhmittelyssä. Käsittelen artikkelissani hieman kattavammin instituutio-keskustelujen teoreettisia ja metodologisia eroavaisuuksia Scottin instituutioiden pylväitä hyväksikäyttäen. Tarkoitukseni on siis testata Scottin jaottelun toimivuutta kuten myös kehitellä Heiskalan mainitsemaa pragmatistisen institutionalismin instituutio-mallia. Lopuksi pyrin vastaamaan kysymykseen, mikä taloussosiologian käsitys instituutioista tulisi olla. 2.2 Keskustelu institutionalismeista Aloitan tarkasteluni taloustieteen uudesta institutionalismista, koska se tuonut keskustelun institutionalismista teoreettisena lähtökohtana uudelleen ajankohtaiseksi yhteiskuntatieteissä. Taloustieteen uudesta institutionalismista on luontevaa lähteä liikkeelle myös siksi, että sen instituutio-käsitys on muita institutionalismeja suppeampi kuten pyrin myöhemmin osoittamaan. 2.2.1 Taloustieteen uusi institutionalismi Taloustieteen valtavirtaa edustava uusklassinen taloustiede perustuu tunnetusti rationaalisen valinnan teorian mukaiseen ajatukseen hyötyä maksimoivasta homo oeconomicus -ihmisestä, jolla on ”täydellinen tai ainakin strukturoitu tietämys kaikista päätöksiinsä vaikuttavista asioista, kognitiivisia rajoitteita ei 32 ole, ja ylipäänsä ihminen toimii kuin sieluton robotti” (Vihanto 2003, 2). Taloustieteellinen uusinstitutionalismi on haastanut uusklassisen teorian oletuksia, mutta se on tehnyt näin hyväksymällä lähtökohdakseen rationaalisen valinnan teorian ja metodologisen individualismin (Eggertsson 1990, xii). Kyse onkin ennemmin uusklassiselle teorialle komplementaarisesta projektista kuin sen keskeisten oletusten korvaamisesta uusilla oletuksilla (Swedberg – Granovetter 2001, 15). DiMaggion ja Powellin (1991, 3) mukaan uusinstitutionalismin yleinen väittämä on ollut se, että yksilöt maksimoivat hyötyään tiettyjen preferenssien pohjalta, mutta he tekevät näin aina kognitiivisten rajoitusten oloissa. Toisaalta uuinstitutionalismi kritisoi uusklassista taloustiedettä siitä, että se olettaa ostamisen ja myymisen olevan kustannuksetonta, jolloin informaation hankinnasta ja taloudellisen toiminnan ympäristön ylläpidosta aiheutuvat kustannukset jäävät huomaamatta (Heiskala 2003a, 17). Toisinaan uusi institutionalismi (New Institutionalism) ja uusinstitutionalismi (Neoinstitutionalism) erotetaan toisistaan, joista jälkimmäinen on tässä erottelussa tiukemmin sitoutunut uusklassisen taloustieteen perustaviin oletuksiin (Eggertsson 1990, 6). Uudella institutionalismilla taas viitataan rajoittunutta rationaalisuutta painottaviin teorioihin (mt., 9). Käytän näitä käsitteitä kuitenkin keskenään synonyymeina, sillä erottelu ei ole vakiintunut käyttöön sen ongelmallisuuden takia. Ongelmallinen se on siksi, että kaikki uusinstitutionaalisen taloustieteen edustajatkaan eivät pidä uusklassisen taloustieteen optimointi-mallia universaalisti sopivana kuvauksena toiminnasta (Rutherford 1996, 1 n3). Taloustieteellisen uusinstitutionalismin käsittelemistä vaikeuttaa se, että siinä ei ole kyse yhtenäisestä teoriarakennelmasta, vaan pikemminkin yrityksestä ottaa instituutiot, organisaatiot ja niiden välinen vuorovaikutus huomioon uusklassisen teorian puitteissa. Taloustieteelliset uusinstitutionalistit ymmärtävät instituutiot regulatiivisina instansseina, eräänlaisina ”pelin sääntöinä”. Taloustieteellisen uusinstitutionalismin tärkeimpinä vaikuttajina mainitaan yleensä Oliver Williamson ja nobelisti Douglass C. North. Williamsonin (1994) transaktiokustannusteoriassa9 instituutioiden oletetaan syntyvän ja säilyvän silloin, kun niiden hyödyt tulevat suuremmiksi kuin niiden ylläpitämisestä aiheutuvat transaktiokustannukset eli niiden neuvottelemisesta, täytäntöönpanosta ja ylläpitämisestä aiheutuvat kustannukset. Instituutiot ovat tällöin jonkinlaisia hallintarakenteita 9 Transaktiokustannukset on suomennettu eri yhteyksissä liiketoiminnan kustannuksiksi, vaihdantakustannuksiksi ja taloustoimikustannuksiksi (Lamberg 1997, 26 n34). Yksinkertaisuuden vuoksi käytän tässä termiä transaktiokustannukset. Myös North käyttää transaktiokustannusten käsitettä, mutta varsinaisesta transaktiokustannustaloustieteestä puhutaan yleensä Williamsonin yhteydessä. Se on Williamsonin (1985, xi) mukaan osa uutta institutionalismia taloustieteessä. 33 (governance structures) eli sosiaalisia sopimuksia, joiden avulla transaktiokustannukset pyritään minimoimaan. Hallintarakenteilla pyritään minimoimaan rationaalisuuden rajallisuudesta ja opportunismista (oman edun tavoittelusta) aiheutuvia ongelmia (Lamberg ym. 1997, 24). Transaktiokustannuksia syntyy aina, kun yksilöt vaihtavat taloudellisten hyödykkeiden omistusoikeuksia ja pitävät yllä eksklusiivisia oikeuksiaan hyödykkeisiin (Eggertsson 1990, 14). Sosiologisesti ajateltuna transaktiokustannuksissa on pohjimmiltaan kyse luottamuksesta ja sen puutteesta, koska asymmetrinen informaatio (sopimuksen toinen osapuoli tietää jotain, mitä toinen ei tiedä) ja epävarmuus päätöksenteossa aiheuttavat transaktiokustannuksia (Nee 1998, 2). Siinä missä Williamson korostaa yksityisten organisaatioiden merkitystä ja yrityskohtaista mikroanalyysia, on Douglass C. North keskittynyt kansantalouden tai jopa maailmantalouden makroanalyysiin ja taloushistorialliseen lähestymistapaan. North on kritisoinut niin sanottua julkisen valinnan koulukuntaa staattisuudesta ja epärealistisesta rationaalisuuskäsityksestä. North pyrkiikin teoriassaan ottamaan huomioon ideologioiden ja uskomusjärjestelmien vaikutuksen ja ymmärtää käyttäytymisen monimutkaisemmaksi kuin uusklassinen talousteoria. Northin mukaan kaikki päätökset ovat kulttuurisidonnaisia ja riippuvaisia yksilöiden kokemusmaailmasta. (Lamberg ym. 1997, 23). Northin tuotannon käsittelemistä vaikeuttaa se, että tuotantonsa alkuvaiheessa North oli sitoutuneempi uusklassisen taloustieteen lähtökohtiin (esim. North – Thomas 1973) kuin myöhemmässä tuotannossaan, jossa ideologioiden ja uskomusjärjestelmien vaikutus on pyritty ottamaan huomioon (North 1990; Denzau – North 1994). Onkin puhuttu ”kahden Northin ongelmasta” (Ankarloo 1999, 148-150). Lambergin ym. (1997, 24) mukaan Northia on kritisoitu siitä, että hänen lähestymistapansa ei käytännössä kuitenkaan eroa rationaalisen valinnan teorian hyödyn maksimoinnista, sillä Northin mallissa ”itsekkäät, rationaaliset toimijat kilpailevat rajoitetuista voimavaroista ja organisaatiot - esimerkiksi taloudelliset painostusjärjestöt - yrittävät institutionaalisen muutoksen kautta saada parempaa tuottoa sijoittamalleen pääomalle” (mt., 24). Sosiologisesta näkökulmasta Northin ongelmana onkin pitäytyminen uusklassisen taloustieteen kehikossa, jossa keskeistä on taloudellisen tuottavuuden analyysi yhteiskunnallisen kokonaisuuden toimintalogiikan analyysin sijaan (Heiskala 2003c, 8). ”Kahden Northin ongelma” ei siis ole niin merkittävä kuin miltä ensisilmäyksellä saattaisi näyttää. Uusinstitutionalismia on kohtalaisen helppo kritisoida sosiologisesta perspektiivistä. Transaktiokustannustaloustieteessä on ensinnäkin havaittavissa luonnontilatyyppistä argumentaatiota, jossa oletetaan alunperin 34 vallinnut uusklassinen puhtaiden markkinoiden tila: ”vapailla markkinoilla hintamekanismi (’markkinavoimat’) ratkaisee kaikki ongelmat” (Lamberg ym. 1997, 29). Transaktiokustannustaloustieteen klassikon Ronald Coasen mukaan onkin puhuttu ”coaselaisesta maailmasta”, jolla tarkoitetaan maailmaa ilman transaktiokustannuksia. Coase oli kuitenkin sitä mieltä, että reaalimaailma ei toimi vapaiden markkinoiden tavoin eli ilman transaktiokustannuksia ja siksi tarvitaankin institutionaalisia ratkaisuja eli lainsäädäntöä takaamaan omistusoikeus sekä oikeudenmukainen resurssien kohdentuminen ja tulonjako (mt., 29). Taloustieteellistä uusinstitutionalismia voidaan syyttää myös funktionalismista, sillä instituutiot ovat uusinstitutionalismin mukaan tehokkaita ratkaisuja markkinoiden epäonnistumisiin. Ideaalitapauksessa markkinat hoitavat talouden toiminnan kitkattomasti, mutta kun näin ei tapahdukaan, tarvitaan instituutioiden apua. Tällainen analyysi on Swedbergin ja Granovetterin (2001, 15) mukaan vain keinotekoisesti sosiaalisen huomioivaa, sillä instituutiot astuvat analyysiin ainoastaan markkinoiden epäonnistuessa. Funktionalismia tällaisessa analyysissa on se, että prosessin lopputuloksella selitetään itse prosessia. Taloustieteellisen uusinstitutionalismin eri muotojen instituutiot ovat Heiskalan (2003a, 17) mukaan yleensä eräänlaisia pelin sääntöjä, joiden olemassaolon ehdoksi nähdään jonkinlainen pakotejärjestelmä, joka rankaisee sääntöjen rikkomisesta. Heiskalan (ma.) esimerkkejä pakotejärjestelmistä ovat valtiollinen sääntely, sulkeminen ulos kaupankäynnistä ja pelien tuomarit. Tämän instituutio-käsityksen lisäksi erilaisia uusinstitutionalismin versioita yhdistää yleensä niin sanotun vanhan institutionalismin kritisoiminen. Useimmiten näissä kritiikeissä on kyse lyhyistä maininnoista, joissa todetaan, että vanha institutionalismi ei kyennyt kehittämään kunnollista teoriaa, ja että se jäi faktojen keräilyn asteelle (Williamson 1994, 78)10. Tässä teoriattomuus- syytöksessä on Rutherfordin (1996, 9) mukaan kyse lähinnä teorian erilaisesta abstraktiotasoista; vanhan institutionalismin holistisempi ja kulttuurisempi perspektiivi saattaa näyttää ”teoriattomalta”, jos sitä vertaa koulutaloustieteen formaaleihin malleihin. Näissä kritiikeissä ei yleensäkään huomioida vanhan institutionalismin keskeisen teoreetikon, Thorstein Veblenin pragmatistista perustaa, joka Kilpisen (2000) mukaan tarjoaa todellisen teoreettisen vaihtoehdon - ei vain uusklassiselle taloustieteelle vaan jopa suurimmalle osalle nykysosiologiaa, kuten pyrin seuraavaksi osoittamaan. 10 Hieman toisenlaisesta kritiikistä, ks. Langlois (1989). Myös nykyisessä taloussosiologiassa on Parsonsin peruina yhdytty vanhan institutionalismin kritiikkeihin (ks. Granovetter 1990, 94). 35 2.2.2 Taloustieteen vanha institutionalismi Vanhan institutionalismin keskeisiksi edustajiksi lasketaan yleensä ainakin Veblen, Wesley Mitchell ja John Commons (Hodgson 1994, 58). Keskityn käsittelemään Thorstein Vebleniä (1857-1929) käyttäen apunani Erkki Kilpisen ja Geoffrey M. Hodgsonin tulkintoja. Veblen ei Kilpisen (2003b, 4 n3) mukaan itse kutsunut lähestymistapaansa institutionalismiksi. Amerikkalaista ”vanhaa” institutionalismia on usein pidetty saksalaisen historismin tai historiallisen koulun jatkeena, ja jos institutionalismi määritellään näin, on perusteltuja syitä väittää, että Veblen ei tällaiseen joukkoon kuulu. Veblen nimittäin väitti, että niin sanotussa metodikiistassa sekä abstraktia ja matemaattista menetelmää kannattaneet marginalistit että historiallisen tutkimustavan tärkeyttä painottaneet historistit olivat yhtä lailla väärässä (Kilpinen 2003c). Institutionalisti-nimike on kuitenkin sen verran vakiintunut Veblenin yhteydessä, että sitä voitaneen edelleenkin käyttää, kunhan vain muistaa Veblenin olleen omalaatuisensa institutionalisti11. Pragmatistisesta filosofiasta ja toimintateoriasta vaikutteita saanut Veblen kritisoi Kilpisen (2000, 192-200) mukaan aikansa uusklassista taloustiedettä neljästä oletuksesta: hedonistisesta yksilökäsityksestä, inhimillisen rationaalisuuden kalkulatiivisesta tulkinnasta, atomistisesta yhteiskuntanäkemyksestä sekä kausaalisuuden ja teleologisuuden asemasta yksilöiden käyttäytymisen ja sosiaalisten prosessien analyysissa. Hedonistinen yksilökäsitys nojaa Veblenin mukaan utilitaristiseen psykologiaan, jossa ihminen oletetaan väärin passiiviseksi olennoksi, jonka ulkoinen negatiivinen tai positiivinen ärsyke ajaa toimimaan. C.S. Peircen pragmatistisen filosofian mukaisesti Veblen ajatteli, että ihminen oli päinvastoin aina jo toiminnassa, jolloin talousteoriankin tulisi lähteä siitä, että jonkinlaista toimintaa tapahtuu jo (mt., 179-181). Toiminnan motiivitkin ilmestyvät tällöin kuvaan kesken jo olemassa olevan toimintaprosessin, jolloin ei ole järkeä tarkastella yksittäisiä tekoja niin kuin uusklassinen taloustiede tekee (Kilpinen 2003b, 10). Myös historiallinen koulukunta oli Veblenin mukaan sidoksissa passiiviseen ihmiskäsitykseen ja niin ollen yhtä lailla väärässä (Kilpinen 2003c). Toisekseen Veblenin mukaan oli väärin tehdä hyötyjä ja haittoja kalkyloivasta rationaalisuudesta jonkinlaista rationaalisuuden paradigmaa, sillä kyse on vain yhdestä rationaalisuuden muodosta, joka kaiken lisäksi on hyvin erityinen tapaus. Laskennallinen rationaalisuus vastaa deduktiivista päättelyä, jossa kaikki tieto on jo premisseissä eikä se siksi tuo mitään uutta 11 Veblen lasketaan myös amerikkalaisen sosiologian klassikoksi. Vanha institutionalismi onkin intellektuaalisesti lähempänä sosiologiaa kuin uutta institutionalismia taloustieteessä (Scott 2001, 2). Veblenin intellektuaalisesta biografiasta, ks. Jorgensen – Jorgensen (1999) ja Edgell (2001); vanhan institutionalismin historiasta, ks. Hodgson (2001, 2003). 36 tietoa maailmaan. Tällaisen päättelyn ja rationaalisuuden lisäksi on kuitenkin myös olemassa tietoa lisäävää järjen käyttöä, jota Peirce kutsui abduktiiviseksi päättelyksi. (Kilpinen 2000, 194–196.) Kolmanneksi atomistinen käsitys yhteiskunnasta, jossa yhteiskunta nähdään ainoastaan yksilöiden summaksi on väärä, koska inhimillinen käyttäytyminen ei tapahdu tyhjiössä, vaan institutionaalisten vaikutusten alaisena. Tätä huomiota voidaan nimittää taloustieteen klassiseksi sosiologiseksi kritiikiksi. Toisin kuin monille muille sosiologian klassikoille, Veblenille pelkkä taloustieteen toimintateorian rikastaminen sosiologisilla lisämausteilla ei kuitenkaan ollut riittävä ratkaisu; koko malli tulee hylätä. (Kilpinen 2000, 197–198.) Neljänneksi taloustiede kieltää yksilöiden toiminnalta jatkuvuuden ja teleologisuuden väittäessään, että jokainen valintatilanne on erillinen tapahtumansa. Teleologista jatkuvuutta taloustiede näkee ainoastaan sosiaalisissa prosesseissa, kuten Adam Smithin teesissä näkymättömästä kädestä, jossa markkinoiden ymmärretään toimivan animistisesti yhteisön jäsenten tietoisista tavoitteista riippumatta (Kilpinen 2003b, 16–17). Tällainen näkemys on Veblenin mukaan väärä, sillä pragmatistisen filosofian mukaisesti toiminnan jatkuvuuden kieltäminen tarkoittaa toiminnan rationaalisuuden kieltämistä. Sosiaaliset prosessit sen sijaan tulisi ymmärtää jatkuvina kumulatiivisina kausaaliprosesseina, joilla ei ole mitään päämäärää. (Kilpinen 2000, 198–200.) Mitä Veblen sitten tarjoaa tilalle? Lyhyesti sanottuna erityistä tavan käsitettä (habit), jolle instituutiot pohjaavat. Veblenille, kuten klassisille pragmatisteille, tapa on järkeen perustuva rutiini (reasoned routine), sillä habituaaliset ja intellektuaaliset aspektit ovat toiminnan kuluessa osittain päällekkäisiä ja vuorovaikutuksessa keskenään. Kyse on siitä, että inhimillinen toiminta tapahtuu aina jossain tietyssä kontekstissa, joka ei ole pysyvä. Tilanteen muuttuessa toimijoiden täytyy sopeutua kontekstin muutoksiin muuttamalla käyttäytymistään eli habituaalistunutta toimintaansa. Veblenin mukaan ei kuitenkaan ole ainoastaan niin, että sosiaaliset instituutiot olisivat yksilöiden habituaalistuneen toiminnan tulosta. Prosessi on pikemminkin kaksisuuntainen: yksilöiden habituaalistunut toiminta tuottaa instituutiot, mutta instituutiot tuottavat yhtälailla yksilöt. Instituutiot ovatkin Veblenille sekä toimintaa rajoittavia että sitä mahdollistavia tekijöitä. Yhteiskunnan produktiossa ja reproduktiossakin on kyse jatkuvasta prosessista, sillä yksilöiden elämäntilanteet ovat jatkuvassa muutoksessa. (Kilpinen 2000, 200– 206.) Suhteessa uusklassiseen taloustieteeseen keskeistä on se, että pragmatismi ymmärtää inhimillisen toiminnan yhtenäisenä ja jatkuvana prosessina, jolloin ei ole mitään järkeä tarkastella yksittäisiä tekoja (Kilpinen 2003a) - kuten 37 yksittäisiä valintatilanteita. Toiminnan motiivit ovat tällöin löydettävissä toimintaprosessin sisäpuolelta, eivätkä sen ulkopuolelta (Kilpinen 2003b). Pragmatismin traditio ja klassinen sosiologia jakavat molemmat Kilpisen (2000, 23) mukaan utilitarismin kritiikin. Erona on kuitenkin se, että siinä missä sosiologia kritisoi utilitarismin ratkaisua toiminnan ja sosiaalisen järjestyksen ongelmaan, pragmatismi kritisoi sen ratkaisua toiminnan ja tietoisuuden ongelmaan. Sosiologia nimittäin esittää utilitarismin kritiikkinä väitteen, jonka mukaan sosiaalinen järjestys ei voi pohjautua ainoastaan omaa etuaan kalkyloivien yksilöiden toiminnalle; tämän lisäksi tarvitaan arvoja ja normeja. Pragmatismi taas esittää, että toiminnan teoretisoiminen kalkylatiivisen rationaalisuutena on ylipäätään väärä lähtökohta. Sosiologia ei ole käsitellyt tätä kysymystä ja siksi utilitaristinen käsitys toiminnasta ja rationaalisuudesta on Kilpisen mukaan hiipinyt kuin varkain oppiaineeseen. Parsonsilainen traditio, johon palaan myöhemmin, on tästä todisteena. Hyväksyessään rationaalisen valinnan teorian oletuksia (mielen ja ruumiin dualismi) traditio siis salakuljettaa utilitarismin teoriaansa. Veblen ei ollut Hodgsonin (2001, 140–141) mukaan kriittinen ainoastaan suhteessa uusklassiseen taloustieteeseen, vaan myös suhteessa Marxiin. Yksilön toiminta ei nimittäin ollut Veblenin mielestä selitettävissä pelkästään sosiaalisesta tai taloudellisesta kontekstista käsin. Veblen toisin sanoen hylkäsi metodologisen individualismin, mutta myös metodologisen holismin. Yksilöllisyyden (individuality) evoluutio on selitettävä kausaalisesti, eikä sitä sovi ottaa annettuna, kuten uusklassisen taloustieteen utilitaristiset ja hedonistiset selitykset tekevät. Yksilön alkuperän selvittäminen johti Veblenin omaksumaan prosessiluonteisen selityksen, jossa huomioidaan sekä biologiset että sosio-ekonomiset seikat. Toisin kuin esimerkiksi Herbert Spencer - ja suuri osa nykypäivän evoluutiopsykologiaa - Veblen ei kuitenkaan redusoinut selityksiä biologiaan. Hodgsonin mukaan on luultavaa, että Veblenin tieteenfilosofiaan vaikutti Conwy Lloyd Morganin näkemys emergenteistä ominaisuuksista. Morgan esitti, että sosio-ekonominen evoluutio tapahtuu emergentillä tasolla, joten se ei ole selitettävissä pelkästään yksilöiden biologisilla ominaisuuksilla, eikä se silloin myöskään edellytä muutoksia ihmisten biologisissa ominaisuuksissa. Veblen lisäsi Morganin näkemykseen institutionalismille keskeisen huomion, jonka mukaan juuri instituutiot ovat valinnan objekteja sosio-ekonomisessa evoluutiossa, vaikkei käyttänytkään emergenttien ominaisuuksien käsitettä. (Hodgson 2001, 140–142.) Veblenin evolutionismin mukaan vaistot tulee huomioida inhimillisen toiminnan selittämisessä. Ne ovat biologis-pohjaisia viettejä tai dispositioita. Veblenille perustavanlaatuisia vaistoja ovat työteliäisyyden vaisto (instinct of 38 workmanship), vanhemmuuden taipumus (parental bent) ja joutilas uteliaisuus (idle curiosity)12. (Hodgson 2003, 146–147.) Keskeisin näistä on työteliäisyyden vaisto, joka tarkoittaa sitä, että työnteko on ihmisluontoon kuuluva välttämätön taipumus. Puhtaana taipumuksena se ei kuitenkaan toteudu käyttäytymisessä ilman ympäristön vaikutusta. Yksilön toiminta on työteliäisyyden vaiston ja kokemuksesta oppimisen eli habituaalistumisen yhteinen tulos. (Kilpinen 2003b, 18-20.) Hodgsonin (2001, 147) mukaan institutionaalisen ja vaistojen evoluution välillä on vuorovaikutusta siten, että muuttuva institutionaalinen ja kulttuurinen ympäristö muokkaa yksilöiden vaistoja. Hodgsonin mukaan pragmatismi tarjoaa myös keinon ylittää marxilaisen materialismin ja saksalaisen historiallisen koulun idealismin välinen dikotomia. Pragmatismi nimittäin osoittaa ideoille materialistisen perustan ja näyttää, kuinka materiaaliset suhteet saavat ilmaisunsa ideoissa. Tämä mahdollistuu ottamalla tavan ja tottumuksen (custom) käsitteet välittäväksi tasoksi. Toisin kuin behavioristis-vaikutteisessa psykologiassa on ajateltu, tapa ei tarkoita käyttäytymistä. Pragmatisteilla ja näihin vaikuttaneille vaistopsykologeilla tapa on ennemminkin alttius tai dispositio, joka ei välttämättä saa ilmaisuaan yksilön kunkinhetkisessä käyttäytymisessä. Mieli ja uskomukset pohjaavat tavoille, jotka taas pohjaavat vaistoille. Näin ollen järkikin on mahdoton ilman sitä edeltäviä habituaalistuneita käsitteitä ja ajattelun rutiineja. Tästä nähdään, että nykypäivän sosiologiaan verrattuna Veblenin ajattelu hylkää mentalistisen toimintakäsityksen pyrkiessään huomioimaan toiminnan ruumiillisen ja biologisen pohjan. Instituutiot eivät vaikuta suoraan yksilöiden päätöksiin, vaan yksilöiden habituaalisiin dispositioihin alaspäin kulkevana kausaationa (downward causation), kuten Hodgson asian ilmaisee. Myös toisensuuntainen kausaatio yksilöistä instituutioihin on mahdollinen, kuitenkaan ottamatta yksilöitä annettuina. Näin voidaan välttää metodologisen holismin vaara, jossa yksilöiden toimijuus redusoituu pelkiksi instituutionaalisiksi vaikutuksiksi. Ajattelun tavat (habits of thought) kun ovat aktiivisia ja luovia elementtejä sopeuduttaessa muuttuviin olosuhteisiin. Hodgson ajattelee Commonsin tavoin, että tottumukset tai käytännöt (custom) ja rutiinit ovat jaettujen tapojen sosiaalisia muodostelmia. Tapa (habit) operoi siis yksilötasolla, kun taas tottumus littyy laajempaan sosiaaliseen ryhmään. Tällöin kulttuuriakin voidaan pitää adaptiivisena systeeminä, joka koostuu habituaalisista elementeistä sekä jonkin sosiaalisen ryhmän jakamista yhteenliittyvistä ja kohtalaisen pysyvistä tottumuksista ja 12 Kilpisen (2000, 189) tulkinnassa joutilas uteliaisuus ei ole oma vaistonsa, vaan työteliäisyyden vaiston variantti. 39 rutiineista. Habituaalisten elementtien taso on aina perustavin, koska sosiaaliset tottumukset pohjautuvat tapoihin. (Hodgson 2001, 288–294.) Kuinka sitten instituutiot muodostuvat ja syntyvät? Tavat ja tottumukset pohjaavat ensinnäkin ihmisluontoon kuuluviin vaistoihin, jotka eivät kuitenkaan realisoidu ilman ympäristön vaikutusta. Tavat ja tottumukset taas mahdollistavat harkinnan muodostumisen ja erilaiset eksplisiittiset säännöt. Instituutiot ovat tällöin Hodgsonin (2001, 295) mukaan ”sosiaalista vuorovaikutusta strukturoivia vakiintuneiden ja uppoutuneiden (embedded) sosiaalisten sääntöjen ja konventioiden kestäviä (durable) järjestelmiä”. Instituutioiden kestävyys johtuu siitä, että ne luovat vakaita odotusrakenteita, joiden avulla tiedämme, mitä odottaa muiden käyttäytymiseltä. Kyse ei kuitenkaan ole pelkistä taloustieteen uuden institutionalismin toimintaa rajoittavista ”pelin säännöistä”, sillä instituutiot myös mahdollistavat toimintaa. Toimija ja rakenne voidaan erottaa toisistaan, mutta molemmat ovat vuorovaikutuksessa ja keskinäisen riippuvuuden suhteessa toisiinsa. (Mt., 295–297.) Siinä missä mieli on biologisten aivojen emergentti ominaisuus, on yhteiskuntarakenne Hodgsonin mukaan yksilöiden toiminnan emergentti ominaisuus. Kumpaakaan ei voida redusoida toiseen, sillä kyse on analyyttisesti erillisistä ilmiöistä, mitä esimerkiksi Giddensin (1984) puhe toimijan ja rakenteen duaalisuudesta ei tavoita13. 2.2.3 Taloussosiologian vanha institutionalismi Sosiologian vanhan institutionalismin teoreetikon tittelin haltijaksi löytyy montakin ehdokasta, vaikka mistään koulukunnasta ei tässä yhteydessä voidakaan puhua. Scott (2001, 9–17) mainitsee muun muassa keskenään niinkin erilaiset teoreetikot kuin Spencerin, Marxin, Durkheimin, Weberin, Parsonsin sekä Bergerin ja Luckmannin. Näiden lisäksi amerikkalaisesta sosiologiasta löytyy myös pragmatistinen perinne, johon aiemmin käsitelty Veblenkin kuuluu, vaikka Vebleniä ei yleensä luokitellakaan sosiologiksi. Tämän perinteen sosiologisina edustajina voidaan sen sijaan pitää William I. Thomasia ja Arthur Bentleytä (Kilpinen 2000). Thomas ja Bentley edustavat kuitenkin pragmatistisen perinteen jatkajina melkolailla samanlaista instituutio-käsitystä kuin Veblen, joten en käsittele heidän teoretisointejaan tässä yhteydessä. 13 Hodgsonin (2003, 32) mukaan jokin ominaisuus on emergentti, jos sen olemassaolo ja luonne riippuu alemman tasoisista entiteeteistä, mutta se ei ole redusoitavissa näihin tai ennustettavissa niistä. Toiminnan ja rakenteen suhde on Hodgsonin teoriassa riippuvainen juuri tällaisesta ”stratifioituneesta ontologiasta” toisin kuin Giddensillä, jolla toiminta ja rakenne ovat saman prosessin eri aspekteja (ne kuvaavat rakenteen duaalisuutta). 40 Émile Durkheim (1982, 28–29) määritteli tunnetusti sosiologian instituutioiden tieteeksi, eikä olekaan vaikea nähdä Durkheimin sosiaalisten faktojen painottamisen tärkeyttä eräänlaisena institutionalismina. Durkheimin käsitys taloudesta on osin samanlainen kuin Veblenin taloustieteelle esittämä kolmas kritiikki: sosiaalisia sopimuksia ei tehdä tyhjiössä, vaan institutionaalisten vaikutusten alaisina. Sosiaalisia sopimuksia siis aina edeltää yhteiskunnan tai yhteisön moraalinen side, eikä niitä näin ollen voida johtaa yksilöiden toiminnasta (ks. Durkheim 1990). Durkheim eroaa Veblenistä kuitenkin siinä, että vaikka Durkheimillakin on tietty ymmärrys tavasta (Camic 1986), se ei ole samalla tavalla refleksiivinen kuin Veblenillä ja pragmatisteilla (Kilpinen 2000, 17). Durkheimin mukaan habituaalisuus ja rationaalisuus nimittäin vuorottelevat toiminnan kuluessa, kun taas pragmatisteilla ne ovat päällekkäisiä ilmiöitä (mt., 93). Durkheim myös olettaa sosiaaliset faktat ontologisesti yksilöistä riippumattomiksi tavalla, jota Veblen ei tee. Taloudelliset instituutiot kun ovat Durkheimin mukaan yleisiä, sosiaalisesti pakottavia ja ulkoisia suhteessa yksilöön. Taloudellisten suhteiden sosiaalinen säätely on funktionaalisesti välttämätöntä, joten taloudelliset instituutiot ilmaisevat tiettyä normatiivista järjestystä. (Beckert 2002, 81.) Talcott Parsons tekee karhunpalveluksen sosiologialle ja leikkaa tavan- käsitteen kokonaan pois sosiologian käsitearsenaalista, jolloin jäljelle jää työskentely sosiaalisilla normeilla ja niihin liittyvillä jäännöskategorioilla (Heiskala 2000, 69). Parsonsia voidaan pitää melko perustellusti jonkinlaisen vanhan institutionalismin edustajana sosiologiassa Parsonsin instituutio- käsityksen vaikutusvaltaisuuden takia. Parsons pyrki tunnetusti laatimaan ”eurooppalaisten klassikoiden työstä USA:n tarpeisiin sovitetun synteesin” (Heiskala 2000, 65). Erik Allardt (1995, 219-220) tiivistää aika hyvin parsonsilaisen käsityksen instituutioista todetessaan, että institutionalisoitumisesta voidaan puhua, kun kolme ehtoa on täytetty. Ensinnäkin suuren osan kulloisenkin sosiaalisen järjestelmän jäsenistä tulee hyväksyä normi tai rooli. Toisekseen normi tai rooli on sisäistynyt, ja kolmanneksi se on pysyvä eli sen noudattaminen on vakiintunutta. Parsonsilaisessa instituutio-käsityksessä instituutioiden nähdään pysyvän pystyssä sanktiouhan lisäksi yhteisöllisen paineen takia ja myös siksi, että toimijat ovat sisäistäneet normit sosialisaatiossa persoonallisuutensa osaksi. Sanktiot ymmärretään tällöin normatiivisena paineena. Jos kuitenkin arvot ja normit on onnistuneesti sisäistetty sosialisaatiossa, eivät instituutiot laisinkaan tarvitse tuekseen sanktioita, sillä toimijat menettelevät sisäistämiensä normatiivisten odotusten mukaisesti myös sellaisissa tilanteissa, joista ulkoinen valvonta puuttuu (Heiskala 2003a, 19). Parsons jätti kuitenkin uusklassiselle taloustieteelle oman pätevyyden alansa, sillä Parsonsin mukaan sosiologia käsittelee toiminnan päämääriä, kun taas taloustiede ottaa 41 päämäärät annettuina ja käsittelee tehokkaita keinoja niiden saavuttamiseksi (Beckert 1996, 823 n80). Voluntaristinen toiminta utilitaristisen kalkyloinnin merkityksessä on kuitenkin mahdollista ainoastaan yhteisöllisten arvo- orientaatioiden johdosta, jotka eivät kalkyloinnin kohteena (Joas 1996, 14). 2.2.4 Taloussosiologian uusi institutionalismi Siinä missä Durkheimin taloustieteeseen ja psykologiaan tiukan eron tekevä instituutioiden tiede sekä etenkin Parsonsin myöhäistuotannon rakennefunktionalismi olivat metodologialtaan holistisia, Mary C Brintonin ja Victor Neen (1998) toimittama uusinstitutionalistista sosiologiaa käsittelevä teos hakee vuoropuhelua taloustieteen kanssa, ja kallistuu sitä hakiessaan metodologisen individualismin suuntaan. Neen (1998, 1) mukaan sosiologinen uusinstitutionalismi tekee pesäeroa Parsonsin funktionalismiin ja - toisin kuin Parsons - pyrkii selittämään instituutioiden olemassaoloa. Toki Parsonskin pyrki selittämään instituutioiden olemassoloa, mutta Parsonsin teoriassa kyse on loppujen lopuksi aina yhteiskunnan arvointegraatioista. Välien selväksi tekeminen Parsonsin perinnön kanssa edellyttää Neen mukaan valintateoreettista lähestymistapaa, joka selittää miten instituutiot tuotetaan ja miten niitä pidetään yllä. Kyse ei kuitenkaan ole uusklassisen teorian hyväksymisestä sellaisenaan vaan kontekstisidonnaisesta rationaalisuudesta, jossa valintoja tehdään instituutioiden asettamissa rajoitteissa (choice-within- constraints). Toimijat toimivat intentionaalisesti epätäydellisen informaation, epätäydellisten mentaalisten mallien ja transaktiokustannusten asettamissa rajoitteissa (ma., 1). Sosiologisen kantaisänsä lähestymistapa löytääkin Max Weberin metodologisesta individualismista ja konteksti-sidonnaisesta rationaalisuudesta. Instituution Nee (ma., 8) määrittelee seuraavasti: Sosiaalisia suhteita säätelevät toisiinsa kietoutuneiden sääntöjen ja normien verkostot eli instituutiot muodostavat ne formaalit ja epäformaalit sosiaaliset rajoitteet, jotka muokkaavat toimijoiden valinta-arsenaalia. Näin ymmärrettynä instituutiot vähentävät ihmisten suhteiden epävarmuutta. Ne määrittelevät legitiimin toiminnan rajat samalla tavoin kuin pelin säännöt määrittävät rakenteen, jossa pelaajat voivat tavoitella strategisia siirtojaan käyttäen pelivälineitä, joilla on tietyt roolit ja statukset. Kuten sitaatin peli-metaforan käytöstä havaitaan, uusinstitutionalistisessa sosiologiassa on aika pitkälti kyse samanlaisesta regulatiivisesta perspektiivistä instituutioihin kuin uusinstitutionalistisessa taloustieteessä - tosin sillä erotuksella, että taloustieteellinen uusinstitutionalismi on 42 sitoutuneempi tarkastelemaan asioita taloudellisen tuottavuuden kautta. Sosiologisessa uusinstitutionalismissa rationaalisuutta rajoittavat pelkkien markkinaympäristöstä juontuvien pakkojen ja psykologisten vajavaisuuksien lisäksi normatiiviset tekijät. Erotuksena Durkheimiin ja Parsonsiin nämä johdetaan valintateoreettisesti. Normit nimittäin ovat Neen (1998, 8) mukaan implisiittisiä tai eksplisiittisiä sääntöjä, jotka ilmentävät yhteisön jäsenten intressejä ja preferenssejä. Ne auttavat ratkaisemaan kollektiivisen toiminnan ongelmia niin, että toimijat hyötyvät yhteistoiminnasta. Normeista tulee sosiaalista pääomaa, jos ne auttavat ratkaisemaan esimerkiksi kuuluisan vangin ongelman tyyppisiä tilanteita, joissa yksilöiden oman edun tavoittelu ei johda kollektiivisesti parhaaseen mahdolliseen lopputulokseen. Normit kertovat, miten tilanteessa tulee toimia, ja auttavat yksilöitä toimimaan yhdessä. Muodollisia ja epämuodollisia normeja valvotaan erilaisin sanktioin ja nämä normeja ylläpitävät mekanismit tuottavat sosiaalisen järjestyksen, eivät normit itsessään (ma., 9). 2.2.5 Organisaatiotutkimuksen institutionalismit Nimenomaisen sosiologisen uuden institutionalismin lisäksi institutionalistiset lähestymistavat ovat olleet vaikutusvaltaisia sosiologisesti inspiroituneessa organisaatiotutkimuksessa. Organisaatiotutkimuksessa uusklassista taloustietedettä on vastannut behavioralistinen näkemys toiminnasta, joka on vastaavalla tavalla olettanut instituutioiden muodostuvan yksilöiden valinnoista. Toisin kuin taloustieteessä, organisaatiotutkimuksen uusinstitutionalismi on hylännyt rationaalisen valinnan teorian ja hakenut pohjaa kulttuurintutkimuksesta ja kognitiota painottavista näkemyksistä. Tällöin instituutioiden ei katsota palautuvan yksilöiden preferensseihin tai valintoihin. (DiMaggio – Powell 1991, 8–9.) Organisaatiotutkimuksessakin siis erotetaan vanha ja uusi institutionalismi. DiMaggion ja Powellin (ma., 12, 15) mukaan organisaatiotutkimuksen vanhassa14 ja uudessa institutionalismissa on keskeisiä eroja, vaikka toisinaan painotetaankin näiden välistä jatkuvuutta. Molemmat teoriarakennelmat lähtevät siitä, että institutionalisoituminen rajoittaa organisaatioiden instrumentaalista rationaalisuutta, mutta siinä missä vanha institutionalismi painotti poliittisten 14 Organisaatiotutkimuksen vanhan institutionalismin ajattelijat eivät itse katsoneet kuuluvansa mihinkään tiettyyn koulukuntaan, joten kyse on retrospektiivisesti hahmotetusta koulukunnasta (Dimaggio & Powell 1991, 12 n9). Vanhan institutionalismin keskeisenä ajattelijana mainitaan yleensä Philip Selznick, joka sai Parsonsin lisäksi vaikutteita Robert Mertonilta (ks. Scott 2001, 22- 24). 43 intressikamppailujen osuutta tässä prosessissa, on uusi institutionalismi painottanut instituutioiden itsestäänselvänä pidettyä luonnetta (ma., 12). Organisaatiotutkimuksen vanha institutionalismi muistuttaa osittain parsonsilaista näkemystä toiminnasta, sillä keskeisenä tekijöinä siinä pidetään arvoja, normeja ja asenteita. Organisaatioiden institutionalisoitumisesta on kyse silloin, kun organisaatiot muodostuvat arvopäämääräksi itsessään. Uusi institutionalismi on tehnyt tähän arvo-orientoituneeseen toimintaan pesäeron korostaessaan sitä, että ennemminkin kuin arvoista ja normeista instituutioissa on kyse itsestäänselvinä pidetystä säännöistä, luokituksista ja ”skripteistä”. DiMaggio ja Powell (1991, 15) näkevät muutoksen taustalla ”kognitiivisen käänteen”, jossa on siirrytty parsonsilaisesta toimintateoriasta etnometodologian ja kognitiivisen psykologian käsitykseen toiminnasta. Kognitiolla DiMaggio ja Powell tarkoittavat sekä järkiperäistä ajattelua (reasoning) että sen esitietoista perustaa (luokitteluja, representaatioita, skeemoja, ”tuotantojärjestelmiä” jne.). DiMaggio ja Powell (ma., 19 n18) kritisoivat Parsonsia toiminnan habituaalisen puolen unohtamisesta. Tässä he tulevat hyvin lähelle pragmatistisesti orientoitunutta institutionalismia. Tosin vaikuttaa siltä, että ainoastaan DiMaggion ja Powellin kognition määritelmä vastaa pragmatismin habituaalisen toiminnan määritelmää - siinä on kyse sekä rationaalisesta että rutiininomaisesta toiminnasta - eikä siis ole niin, että juuri habituaalisen toiminnan määritelmät kohtaisivat toisensa. DiMaggio ja Powell tuntuvat nimittäin varaavan tavalle lähinnä rutiinin roolin. Kognitiiviset painotukset tulevatkin organisaatiotutkimukseen muualta kuin pragmatismista, nimittäin Herbert Simonilta ja James Marchilta, jotka painottivat organisationaalisten rutiinien kykyä vähentää epävarmuutta, joka seuraa - contra uusklassinen taloustiede - kognitiivisista rajoitteista (DiMaggio – Powell 1991, 18–19). Simonin käsityksen mukaan tapa ohjaa toimijan huomion kunkin tilanteen tiettyihin piirteisiin toisten piirteiden kustannuksella. Varsinaista toimintateoreettista pohjaa uudelle institutionalismille DiMaggio ja Powell (1991, 19–22) hakevat Garfinkelin etnometodologiasta sekä Bergerin ja Luckmannin fenomenologisesta sosiologiasta. Etnometodologian keskeisenä antina he pitävät kognition näkemistä pääosin ei-tietoisena, rutiineihin ja ”praktiseen järkeen” nojaavana prosessina. Bergerin ja Luckmannin (1995) käsitys instituutioista habituaalisen toiminnan vastavuoroisina tyypittelyinä muistuttaa DiMaggion ja Powellin (1991, 21) mukaan Parsonsin näkemystä institutionalisoituneista rooleista, mutta erona on kuitenkin se, että Bergerin ja Luckmannin analyysi operoi kognition tasolla, kun taas Parsons painottaa evaluatiivista ja affektiivista aspektia rooliodotusten sisäistämisessä. 44 Tyydyttävimmän toimintateoreettisen pohjan uudelle institutionalismille DiMaggion ja Powellin mukaan tarjoaa kuitenkin Bourdieun teoria habituksesta, vaikka keskeisimmät teoreettiset vaikutteensa organisaatiotutkimuksen uusi institutionalismi näyttääkin yleensä ottavan etnometodologiasta ja fenomenologisesta sosiologiasta (ks. Scott 2001, 37– 41). DiMaggio ja Powell (1991, 25–26) korostavat habitus-käsitteen avulla sitä, että institutionalisoitumisessa on kyse toiminnan itsestäänselvinä pidetyistä elementeistä (Bourdieun ”doksa”), sosiaalisista luokituksista, praktisesta tietoisuudesta (”tiedosta ilman käsitteitä”) sekä yhteiskuntarakenteiden tilanteellisesta ja ruumiillisesta reproduktiosta. Tästä saadaankin haluttaessa läheinen yhteys taloustieteen vanhaan institutionalismiin. Kilpisen (2000, 21) mukaan Bourdieun habitus nimittäin on lähellä pragmatistien tavan käsitettä ja niin ollen myös vebleniläistä institutionalismia. Molempien käsitteiden tarkoituksena on yhdistää sosiaalinen rakenne ja yksilön toimijuus, ruumiillinen ja mentaalinen sekä toiminnan tietoinen ja ei-tietoinen aspekti. Kilpinen kuitenkin toteaa, että Bourdieun habitus ei ole yhtä refleksiivinen käsite kuin pragmatistien tapa, eikä Bourdieulla myöskään ole paljonkaan sanottavaa habituksen rajoista eli tilanteista, joissa tapa tai habitus joutuu muuttuvan objektiivisen tilanteen takia kriisiin. Pragmatisteilla tavat muuttuvat tällaisissa tilanteissa. ”Katastrofi tai shokki luo tapojen muuttumisen tavan”, kuten Kilpinen toteaa Peirceä siteeraten. (Mt., 21–22.) 2.3 Instituutioiden pylväät Eri institutionalismien keskusteluja yhdistää luonnollisestikin instituutioiden nostaminen keskiöön. Ei kuitenkaan ole laisinkaan selvää, mitä instituutiolla kulloinkin tarkoitetaan. Richard Scott (2001, 51–58) on puhunut instituutioiden kolmesta pilarista tai pylväästä. Scottin (mt., 40) mukaan instituutiot koostuvat regulatiivisista, normatiivisista ja kulttuuris- kognitiivisista elementeistä tai pylväistä. Pylväiden sijasta on mahdollista puhua myös erilaisista institutionalismeista, kuten Heiskala (2003a, 17–20) on Scottia mukaillen tehnyt. Ismeistä puhuminen korostaa kuitenkin eri instituutio-näkemysten erilaisuutta tavalla, jota puhe pylväistä ei tee. Scottin mukaan pylväät nimittäin ovat toisilleen komplementaarisia ja ne muodostavat jatkumon, jossa liikutaan tietoisesta esitietoiseen ja legaalisesti sanktioidusta säännöistä itsestäänselvinä pidettyihin kulttuurisiin skripteihin. Regulatiivinen näkemys instituutioista painottaa instituutioiden toimintaa rajoittavia ja regularisoivia aspekteja. Tällöin on kyse jonkinlaisista ”pelin säännöistä” ja näiden sääntöjen ylläpitämiseksi tarvittavista 45 sanktiomekanismeista. Regulatiivinen pylväs sijoittuu instituutioiden jatkumolla legaalisesti sanktioitujen sääntöjen päähän, vaikkei kyse tarvitsekaan olla muodollisista säännöistä, kuten laeista. (Scott 2001, 51–54.) Normatiivisessa instituutioiden pylväässä instituutioiden nähdään pysyvän pystyssä sanktiouhan lisäksi yhteisöllisen paineen takia ja myös siksi, että toimijat ovat sisäistäneet normit sosialisaatiossa persoonallisuutensa osaksi. Sanktiot ymmärretään tällöin normatiivisena paineena. Jos kuitenkin arvot ja normit on onnistuneesti sisäistetty sosialisaatiossa, eivät instituutiot laisinkaan tarvitse tuekseen sanktioita, sillä toimijat menettelevät sisäistämiensä normatiivisten odotusten mukaisesti myös sellaisissa tilanteissa, joista ulkoinen valvonta puuttuu (Heiskala 2003a, 19). Osa normeista ja arvoista pätee kuitenkin vain tiettyihin toimijoihin ja toimijapositioihin. Tällöin on kyse erilaisista rooleista. Scottin (2001, 55) mukaan normatiiviset järjestelmät eivät ainoastaan rajoita sosiaalista käyttäytymistä, sillä ne myös mahdollistavat sitä delegoimalla toimijoille velvollisuuksien lisäksi oikeuksia. Toisin kuin regulatiivinen näkemys, normatiivinen näkemys sulkee pois sellaiset säännöt, jotka eivät ole legitiimejä (Heiskala 2003b, 4). Kolmantena instituutioiden muotona Scott (2001, 57–58) erottaa kulttuuris- kognitiivisen pylvään, jossa instituutiot ymmärretään kulttuurisina merkityskehyksinä, joiden ylläpitämisessä sanktioilla ei ole juurikaan roolia. Kyse on jaetuista ja itsestäänselvyyksinä pidetyistä käsityksistä ja merkityskehyksistä, jotka konstituoivat sosiaalisen todellisuuden. Ne eivät ole pelkästään subjektiivisia uskomuksia, sillä institutionalisoitumista tapahtuu silloin, kun merkitysjärjestelmät nähdään objektiivisina ja toimijoiden ulkopuolella olevana. Keskeiseksi kiinnostuksen kohteeksi tällöin tulevat rutiineissa toistetut itsestäänselvyydet. (Scott 2001, 57–58.) Heiskala (2003a, 18-20) lisää Scottin erotteluun neljännen pylvään15, jota Heiskala kutsuu pragmatistiseksi institutionalismiksi. Jos kuitenkin halutaan korostaa instituutio-näkemysten komplementaarisuutta, kuten Heiskala (2003b, 4–5) on itsekin tehnyt, voi edelleenkin olla hedelmällisempää pitäytyä puheessa instituutioiden pylväistä. Tätä neljättä pylvästä voitaisiin paremman puutteessa nimittää habituaaliseksi pylvääksi, koska sen keskeinen käsite on taloustieteen vebleniläisen institutionalismin tapa ja tavan kautta habitualisoituminen. Myös Scott (2001, 2–5) tunnistaa vanhan institutionalismin olemassaolon, mutta se tuntuu olevan hänelle lähinnä oppihistoriallisesti kiinnostava kohde, eikä Scott huomioi erityistä tavan- käsitettä, jonka osa vanhan institutionalismin edustajista otti pragmatistisesta 15 Scottkin (2001, 58 n4) mainitsee alaviitteessä neljännen kandidaatin institutionaaliseksi pylvääksi, jota hän kutsuu kateettiseksi tai emotionaaliseksi pylvääksi. Tämä jää Scottin käsittelyssä kuitenkin pelkäksi maininnaksi. 46 filosofiasta. Instituutioiden habituaalisen pylvään kautta päädytään hyvin laajaan instituution määritelmään, jossa ”instituutioita ovat paitsi yhteiset merkityskehykset myös sellaiset tavat, joiden olemassaolosta toimijat eivät välttämättä lainkaan ole tietoisia” kulttuuris-kognitiivisella tasolla (Heiskala 2003, 19). Esitän taulukossa 1. tiivistettynä instituutioiden kolmen pylvään keskeiset toimintaperiaatteet sellaisena kuin Scott (2001, 52) ne esittää (termien suomennokset Heiskala 2003a, 19). Olen lisännyt Scottin taulukkoon ehdotukseni neljänneksi pylvääksi (habituaalinen pylväs) ja alustavan arvioni sen toimintaperiaatteista sekä eri institutionalismien paikat. Taulukko 1. Instituutioiden neljä pylvästä, niiden toimintaperiaatteet ja eri institutionalismit. Regulatiivinen Normatiivinen Kulttuuris-kognitiivinen Habituaalinen Institutionalis- mi Taloustieteen ja sosiologian uudet institutionalismit Sosiologian ja organisaatio- tutkimuksen vanhat institutionalismit Organisaatio- tutkimuksen uusi institutionalismi Taloustieteen vanha institutiona- lismi (Veblen) Yhteistyön perusta Tarkoituksen- mukaisuus16 Sosiaalinen velvollisuus Itsestäänselvyys, yhteinen merkitys Yhteiset tiedostamat- tomat merkitykset Järjestyksen perusta Regulatiiviset säännöt Sitovat odotukset Konstitutiivinen skeema Itsestäänselvät habituaaliset toimintatavat Mekanismit Pakottava Normatiivinen Mimeettinen Habituaalinen Logiikka Instrumentaali-suus Sopivaisuus Ortodoksia Pragmaatti- suus Indikaattorit Säännöt, lait, sanktiot Kelpuutus, valtuutus Yhteiset uskomukset, yhteiset toimintalogiikat (Ajattelun) tavat, tiedostamat- tomat kulttuuriset faktat Legitimaatio- perusta Laillinen sanktiointi Moraalinen hallinta Ymmärrettävyys, tunnistettavuus, perustuminen kulttuuriin Toimivuus, toiminnan sujuvuus Eri institutionalismit istuvat Scottin ja Heiskalan oletusten mukaisesti melkoisen vaivatta instituutioiden pylväiden jaotteluun. Scottin (2001, 52) 16 Tarkoituksenmukaisuus tavoittaa mielestäni paremmin sanan expedience merkityksen tässä yhteydessä kuin Heiskalan (2003a, 19) suomennos, joka on toimivuus. 47 mukaan taloustieteen uusinstitutionalismi nojaa instituutioiden regulatiiviseen pylvääseen, jossa instituutioilla tarkoitetaan pelin sääntöjä, joiden takana on jokin pakotejärjestelmä. Lisäisin tähän kuitenkin huomion, jonka mukaan myös Brintonin ja Neen (1998) sosiologinen uusinstitutionalismi edustaa lähinnä regulatiivista instituutio-käsitystä. Normatiivinen käsitys instituutioista taas on ollut enemmän tai vähemmän klassisten sosiologien heinää aina Durkheimista Parsonsiin. Suomalaisessa sosiologiassa Allardt (1995) on hyvä esimerkki normatiivisesta instituutiokäsityksestä. Organisaatiotutkimuksessa normatiivista kantaa on edustanut vanha institutionalismi (esim. Philip Selznick) (Scott 2001, 55). Organisaatiotutkimuksen uusi institutionalismi taas on omaksunut kulttuuris-kognitiivisen instituutio-näkemyksen, joka on saanut vaikutteita muun muassa Bergerin ja Luckmannin (1995) fenomenologisesta tiedonsosiologiasta ja myös antropologiasta (esim. Douglas 1986). Scott (2001, 57) väittää instituutioiden kulttuuris-kognitiivisen ulottuvuuden erityisesti erottavan kaiken sosiologis-vaikutteisen uuden institutionalismin muista institutionalismeista. Väitteessä on toki perää, sillä kulttuuris- kognitiivista ulottuvuutta on vaikea löytää taloustieteen uudesta institutionalismista ainakaan samassa muodossa kuin sosiologiasta. Brintonin ja Neen nimenomaisesti sosiologista uusinstitutionalismia käsittelevä teos The New Institutionalism in Sociology (1998) kuitenkin edustaa lähinnä regulatiivista näkemystä, joten kulttuuris-kognitiivisen ulottuvuuden painottaminen ei vastoin Scottin oletusta luonnehdi kaikkea sosiologis- vaikutteista uusinstitutionalismia, vaikka se onkin ollut vaikutusvaltainen koko sosiologian kentällä erilaisten sosiaalisten konstruktionismien muodossa. Scottia vastaan puhuu myös se seikka, että organisaatiotutkimuksen uuden institutionalismin sisälläkin on pyrkimyksiä käsitteellistää merkityksiä, joilla ei ole tietoista tulkintaa, kuten huomasimme DiMaggion ja Powellin (1991) nojaamisesta Bourdieu’n habitukseen. Parsonsin jälkeisestä sosiologiasta on pitkälti puuttunut amerikkalaisen pragmatismin mukainen instituutio-käsitys, jota nimitän instituutioiden habituaaliseksi pylvääksi ja joka luonnehtii Veblenin vanhaa institutionalismia. Kulttuuris-kognitiiviselle tasolle jäämisen ongelmana on se, että kaikki merkitys ymmärretään siinä tietona. Heiskalan (2000, 104–105) mukaan esimerkiksi Berger ja Luckmann (1995) yrittävät käsitellä reflektoimattomia tajunnan huomion kiinnittymisen akteja ja tapoja, joilla ei ole tulkintaa, mutta heidän teoreettinen käsitteistönsä on rajattu niin, ettei se mahdollista näiden ilmiöiden tarkastelua17. Tämä johtuu siitä, että heidän 17 Tällaista menettelyä voidaan Parsonsia mukaillen kutsua työskentelyksi jäännöskategorioilla (Heiskala 2000, 105). 48 rajauksessaan merkitys samaistuu reflektiiviseen intentionaaliseen aktiin. Näin käy myös tavan käsitteelle, jonka Berger ja Luckmann johtavat tietoisesta toiminnasta. Fenomenologisessa sosiologiassa toimijaa pidetäänkin Heiskalan (mt., 19) mukaan merkitystulkinnoissaan itselleen läpinäkyvänä - samoin kuin uusklassisessa taloustieteessä, jossa toimijalla ei ole kognitiivisia rajoitteita, kuten Heiskalan tulkintaan voisi lisätä. Edellä käsitellyt pragmatistit ajattelivat toisin, sillä heidän mukaansa ”tietoista tulkintaa edeltävien tapojen olemassaolo on perusta, joilta tapoja koskevat uskomukset ja tieto voivat kehittyä” (mt., 110). Kyse ei Kilpisen (2000) mukaan kuitenkaan ollut kuolleesta rutiinista, kuten edellä nähtiin. Heiskala (2003b, 4) on toisaalla esittänyt, että Scottin erittelemät instituutioiden pylväät vastaavat erilaisia näkemyksiä yhteiskuntarakenteesta. Regulatiivinen pylväs vastaa ensinnäkin näkemystä, jossa yhteiskuntarakenne nähdään organisationaalisena rakenteena ja instituutioiden muodostamana kokonaisuutena. Normatiivinen pylväs taas vastaa yhteiskunnan normatiivista rakennetta, jossa rakenteella viitataan vastavuoroisten odotusten kokonaisuuden muodostamiin instituutioihin. Kolmantena yhteiskuntarakenteen mallina Heiskala erottaa yhteiskunnan kulttuurisen rakenteen, joka ei kuitenkaan täysin vastaa instituutioiden kolmatta eli kulttuuris-kognitiivista pylvästä. Kulttuurinen rakenne pitää sisällään mentaaliset paradigmat, kognitiiviset kehykset ja tulkinnan tavat. Kulttuurinen rakenne voidaan määritellä niin laajasti, että se sisällyttää itseensä sekä yhteiskunnan organisationaalisen että normatiivisen rakenteen. Tiputtamalla kognitiivis-päätteen käsitteestä Heiskala yrittää välttää kulttuuris-kognitiivisen pylvään aiemmin mainitun ongelman eli merkityksen samaistumisen tietoon. (Heiskala 2003b, 4 n3.) Ongelma voidaan siis yrittää välttää muutenkin kuin lisäämällä jaotteluun erillinen pragmatistinen institutionalismi tai instituutioiden habituaalinen pylväs. Pitäydyn itse kuitenkin tässä lisäyksessä siksi, että se palvelee tutkimukseni analyyttisia tarkoituksia. Erottamalla habituaalinen instituutioiden pylväs omaksi pylvääkseen voidaan nimittäin helpommin havaita taloustieteen vanhan institutionalismin erityisen instituutio-käsityksen ja toimintateorian ero suhteessa organisaatiotutkimuksen uuteen institutionalismiin ja sosiologian erilaisiin sosiaalisiin konstruktionismeihin, joiden instituutio-käsitykset asettuvat kaikki Heiskalan kulttuurisen rakenteen alle. Heiskalakaan ei suoraan kiistä habituaalisen pylvään mahdollisuutta, vaan toteaa, että oman tutkimuksensa kontekstissa kulttuurisen rakenteen ja kulttuuris-kognitiivisen pylvään eroilla ei ole merkitystä. Näin ei kuitenkaan välttämättä ole kaikissa tutkimuksellisissa konteksteissa. Instituutioiden habituaalisen pylvään olemassaolo - vaikkakaan ei suoranaisesti tällä käsitteellä - on viime aikoina huomioitu lähinnä taloustieteellisen vanhan 49 institutionalismin uudelleenlämmittelyissä (esim. Hodgson 2001; 2003). Harvoja sosiologisia poikkeuksia ovat Joas (1996) ja Kilpinen (2000, 2003b), mutta heidän teoretisointinsa eivät aivan suoranaisesti kytkeydy keskusteluun instituutioista (ks. tosin Kilpinen 2003a). 2.4 Instituutioiden taustaoletukset ja integroidun instituutiomallin mahdollisuus Heiskalan (2003, 20) mukaan eri institutionalismien välillä voidaan käydä ankaraakin kamppailua, mutta usein on kiistelyn sijaan järkevämpää valita ”kulloiseenkin tutkimustehtävään se määritelmä, joka parhaiten soveltuu käyttöön”. Tällöin tietyn määritelmän valittuaan tutkijan ei tarvitse pitää instituutioiden muita pylväitä virheellisinä näkemyksinä, vaan ainoastaan kyseiseen tutkimustehtävään sopimattomina. Scott (2001, 69) taas huomauttaa, että erilaiset instituutio-näkemykset olisi periaatteessa mahdollista yhdistää integroiduksi instituutio-malliksi, mutta hyödyllisempää on kuitenkin eri institutionalismien erilaisten taustaoletusten huomioiminen. Teoreettiset ja metodologiset taustaoletukset on luonnollisestikin syytä huomioida, mutta tämä ei kuitenkaan sulje pois kaikki instituutioiden pylväät sisältävän instituutio-mallin mahdollisuutta. Tällaisessa mallissa liikuttaisiin suppeammasta määritelmästä kattavampaan liikuttaessa regulatiivisesta pylväästä habituaalisen pylvään suuntaan niin, että seuraava pylväs aina sisällyttäisi edellisen pylvään itseensä. Heiskala taas on varmasti oikeassa siinä, että turhaa kiistelyä tulee välttää, mikä onnistuneekin tiettyyn pisteeseen asti tuntemalla institutionalismien taustaoletukset. Tämä ei kuitenkaan aseta mitään periaatteellisia esteitä instituutioiden pylväistä muodostuvalle, integroidun instituutio-mallin samanaikaiselle rakentamiselle. Ongelma on ennemminkin siinä, että institutionalismien taustaoletukset ovat tietyin paikoin keskenään ristiriidassa. Koska instituutioiden habituaalinen pylväs on kaikista kattavin, ja eräällä tavalla muut pylväät itseensäsisällyttävä, esitän että se tulee asettaa mittapuuksi, joihin muut instituutioiden pylväät taustaoletuksineen tulee sopeuttaa - siis mikäli integroitua instituutioiden mallia pyritään rakentamaan. Tämä tarkoittaa alustavasti hahmotettuna sitä, että ensinnäkin taloustieteen uusinstitutionalismin regulatiivinen pylväs pitää nähdä siten, että laskelmoivaa rationaalisuutta ei enää pidetä rationaalisen toiminnan perusmuotona, vaan sen erityistapauksena. Hedonistinen eli passiivinen ihmiskäsitys sekä atomistinen näkemys yhteiskunnasta tulee myös hylätä - sortumatta kuitenkaan suoranaiseen metodologiseen holismiin. Sosiologian perspektiivistä myös taloustieteen taloudellisen tuottavuuden analyysin pitää väistyä 50 yhteiskunnallisen kokonaisuuden toimintalogiikan selvittämisen tieltä (ks. Heiskala 2003c, 8). Normatiivisen pylvään taustaoletuksissa taas on ongelmana se, että se ei tavoita sosiaalista todellisuutta konstituoivia jaettuja ja itsestäänselvyyksinä pidettyjä käsityksiä ja merkityskehyksiä, joiden ylläpitämisessä suoranaisilla sanktioilla ei ole roolia. Kulttuuris-kognitiivista pylvästä hahmotettaessa tulee huomioida se, että kaikki merkitys ei palaudu tietoon. Tämän huomion tehtyämme astumme habituaalistuneen toiminnan alueelle, jonka varassa toiminta normaalisti etenee - kunnes ajaudumme kriisiin ympäristön laittaessa hanttiin. Eri instituutio-näkemyksistä on siis mahdollista edetä asteittain toisiinsa ja rakentaa synteesi, joka käsittelee tyydyttävästi toiminnan ja rakenteen kysymyksiä (vrt. Heiskala 2000, 15). 51 Kirjallisuus Allardt, Erik (1995) Sosiologia I. WSOY: Juva. Ankarloo, Daniel (1999) “Institutions”, What is in a Word? A Critique of the New Institutional Economics. Department of Economic History: Lund University. Lund Studies in Economic History 11. Aranson, Peter H. (1998) The New Institutional Analysis of Politics. Journal of Institutional and Theoretical Economics, Vol. 154, No.4, 744– 753. Beckert, Jens (1996) What is Sociological about Economic Sociology? Uncertainty and the Embeddedness of Economic Action. Theory and Society, Vol. 25, No. 6, 803–840. Beckert, Jens (2002) Beyond the Market. The Social Foundations of Efficiency. Princeton University Press: Princeton and Oxford. Berger, Peter L. – Luckmann, Thomas (1995) Todellisuuden sosiaalinen rakentuminen. Tiedonsosiologinen tutkielma. Gaudeamus: Helsinki. Brinton, Mary C. – Nee, Victor (eds.) (1998) The New Institutionalism in Sociology. Russel Sage Foundation: New York. Denzau, Arthur T. – North Douglass C. (1994) Shared Mental Models: Ideologies and Institutions. Kyklos, Vol. 47., No. 1, 3–31. DiMaggio, Paul J – Powell, Walter W. (1991) Introduction. Teoksessa: The New Institutionalism in Organizational Analysis, toim. Paul J. DiMaggio ja Walter W. Powell, 1–38. University of Chicago Press: Chicago and London. Douglas, Mary (1986) How Institutions Think. Syracuse University Press: New York:. Durkheim, Émile (1982) Sosiologian metodisäännöt. Tammi: Helsinki. Durkheim, Émile (1990) Sosiaalisesta työnjaosta. Gaudeamus: Helsinki. Edgell, Stephen (2001) Veblen in Perspective. His Life and Thought. M.E. Sharpe: New York. Eggertsson, Thráinn (1990) Economic Behavior and Institutions. Cambridge University Press: Cambridge. Ensminger, Jean (1998) Anthropology and the New Institutionalism. Journal of Institutional and Theoretical Economics, Vol. 154, No. 4, 774– 789. Giddens, Anthony (1984) The Constitution of Society. Outline of the Theory of Structuration. Polity Press: Cambridge. 52 Granovetter, Mark (1990) The Old and the New Economic Sociology: A History and an Agenda. Teoksessa: Beyond the Marketplace. Rethinking Economy and Society, toim. Roger Friedland ja A.F. Robertson, 89–112. Aldine de Gruyter: New York. Heiskala, Risto (1997) Society as Semiosis. Neostructuralist theory of culture and society. Research Reports No. 231. Department of Sociology, University of Helsinki. Heiskala, Risto (2000) Toiminta, tapa ja rakenne. Kohti konstruktionistista synteesiä yhteiskuntateoriassa. Gaudeamus: Helsinki. Heiskala, Risto (2003a) Instituutiot, sosiaaliset innovaatiot ja yhteiskunnan rakenteellinen muutos. Teoksessa Yhteiskunnallinen muutos, toim. Harri Melin ja Jouko Nikula, 15–27. Vastapaino: Tampere. Heiskala, Risto (2003b) Some Basic Concepts for the Study of Social Innovation. Working Paper for the SAEP Workshop, 12–13 June, Majvik. Heiskala, Risto (2003c) Uusi taloussosiologia: globalisaatio, instituutiot ja kulttuuri. Teoksessa Talous ja yhteiskuntateoria, toim. Risto Heiskala ja Akseli Virtanen. Gaudeamus. (Ilmestyy. Viittaukset käsikirjoituksen sivunumeroihin.) Hodgson, Geoffrey M. (1994) The Return of Institutional Economics. Teoksessa The Handbook of Economic Sociology, toim. Neil J. Smelser ja Richard Swedberg, 58–76. Princeton University Press: Princeton. Hodgson, Geoffrey M. (2001) How Economics Forgot History. The Problem of Historical Specifity in Social Science. Routledge: London and New York. Hodgson, Geoffrey M. (2003) Reconstructing Institutional Economics: Evolution, Agency and Structure in American Institutionalism. (Ilmestyy. Viittaukset käsikirjoituksen sivunumeroihin.) Joas, Hans (1996) The Creativity of Action. Polity Press: Cambridge. Jorgensen, Elizabeth W. – Jorgensen, Henry I. (1999): Thorstein Veblen. Victorian Firebrand. M.E. Sharpe: New York. Kilpinen, Erkki (2000) The Enormous Fly-Wheel of Society. Pragmatism’s Habitual Conception of Action and Social Theory. Department of Sociology: University of Helsinki. Research Raports No. 235. Kilpinen, Erkki (2003a) Does Pragmatism Imply Institutionalism? Journal of Economic Issues, Vol. 41, No. 2. (Ilmestyy. Viittaukset käsikirjoituksen sivunumeroihin.) Kilpinen, Erkki (2003b) Thorstein Veblen ja talousteorian evolutiivinen kritiikki. Teoksessa: Talous ja yhteiskuntateoria, toim. Risto 53 Heiskala ja Akseli Virtanen. Gaudeamus. (Ilmestyy. Viittaukset käsikirjoituksen sivunumeroihin.) Kilpinen, Erkki (2003c) How to Fight the ’Methodenstreit’? Veblen and Weber on Economics, Psychology and Action. Julkaisematon esitelmä. Langlois, Richard N. (1989) What Was Wrong With the Old Institutional Economics (and What is Still Wrong With the New)? Review of Political Economy, Vol. 1, No. 3, 270–298. Lamberg, Juha-Antti – Ojala, Jari – Eloranta, Jari (1997) Uusinstitutionalismi ja taoushistoria. Kollektiivisen valinnan ja liiketoiminnan kustannusten problematiikka. Teoksessa: Uusi institutionaalinen taloushistoria. Johdanto tutkimukseen, toim. Juha-Antti Lamberg ja Jari Ojala, 15–47. Atena Kustannus: Jyväskylä. Nee, Victor (1998) The Sources of the New Institutionalism. Teoksessa: The New Institutionalism in Sociology, toim. Mary C. Brinton ja Victor Nee, 1–16. Russel Sage Foundation: New York. North, Douglass C. (1990) Institutions, Institutional Change and Economic Performance. Cambridge University Press: Cambrigde. North, Douglass C. – Thomas, Robert Paul (1973) The Rise of the Western World. A New Economic History. Cambridge University Press: Cambrigde. Raumolin, Jussi (1995) Ympäristötaloustieteen koulukuntia ja suuntauksia. Teoksessa: Riskiyhteiskunnan talous. Suomen talouden ekologinen modernisaatio, toim. Ilmo Massa ja Ossi Rahkonen, 41–96. Helsinki: Gaudeamus. Rutherford, Malcolm (1996) Institutions in Economics. The Old and The New Institutionalism. Cambridge University Press: Cambridge. Saari, Juho (toim.) (2003) Instituutiot ja sosiaalipolitiikka. Johdatus institutionaalisen muutoksen tutkimukseen. Sosiaali- ja terveysturvan keskusliitto ry: Helsinki. Scott, Richard W. (2001) Institutions and Organizations. 2nd Edition. Sage:Thousand Oaks. Swedberg, Richard - Granovetter, Mark (2001) Introduction to the Second Edition. Teoksessa: The Sociology of Economic Life, toim. Richard Swedberg ja Mark Granovetter, 1–28. Westview Press: Boulder. Vihanto, Martti (2003) Uusinstitutionaalinen taloustiede ja sen kehittämissuuntia. Teoksessa: Talous ja yhteiskuntateoria, toim. Risto Heiskala ja Akseli Virtanen. Gaudeamus. (Ilmestyy. Viittaukset käsikirjoituksen sivunumeroihin.) 54 Williamson, Oliver E. (1985) The Economic Institutions of Capitalism. Firms, Markets, Relational Contracting. The Free Press: New York and London. Williamson, Oliver E. (1994) Transaction Cost Economics and Organization Theory. Teoksessa: The Handbook of Economic Sociology, toim. Neil J. Smelser ja Richard Swedberg, 77–107. Princeton University Press: Princeton. 55 3 PERSPECTIVES ON CONSUMPTION AND ENVIRONMENT Leena Haanpää Abstract Consumption and consumerism are problematic fields of research when it comes to environmental questions. On the one hand, there have been attempts to increase environmentally responsible or green consumption by educating consumers to make environmentally benign consumption choices in different phases of consumer behaviour, e.g. to buy green products and to recycle them. On the other hand, it has been asked, whether consumption in general can be environmentally sound in any aspect because of its ecological impacts. The main interest of this paper is to examine the importance of micro level’s consumption, consumers’ role in environmental questions. Perspective on this subject is given by evaluating the core theme of ecological modernisation theory, social change when it comes to consumer’s role in bringing about environmentally induced social change. Thus, consumers’ role is elaborated by monitoring the notion mainly through the ideas developed in the ecological modernisation theory and also drawing linkages to environmental sociology. KEYWORDS: green consumption, green consumers, social change, ecological modernisation. 3.1 Introduction The general “greening” of opinion climate of western society during the last three decades of 20th century has led to a holistic approach within the thinking of economic activity. According to that view environment is seen as the foundation of the society (Peattie 1999, 137), which is supported by social transformations. Social transformations involve not only physical changes concerning environmental disruptions and material flows but also social and institutional improvements (Mol – Sonnenfeld 2000, 5–6; Sonnenfeld – Mol 2002a; Spaargaren 1999). These social changes are in the core of ecological 56 modernisation theory. Mol and Sonnenfeld (2000, 6) group the transformations in following clusters: changing role of science and technology, increasing importance of market dynamics and economic agents, transformations in the role of the nation-state, modifications in the position, role and ideology of social movements and lastly, changing discursive practices and emerging new ideologies. The second transformation cluster, market dynamics and economic agents, is connected to environmental, green consumption and consumerism. With the terms “market dynamics” and “economic agents” the authors refer to different kind of actors having many transporting roles, which affect to the ecological reform and change. Such actors are e.g. producers, consumers and credit institutions. Debate among scholars has emerged about whether social changes have remained just a theoretic speculation without having actually taken place (Mol and Sonnenfeld 2000, 6). The so called green wakening of societies and consumers to environmental questions and to the recognition of a collective problem has not led to considerable changes of e.g. consumption manners (Olney & Bryce 1991, 695). The improving changes of consumption habits would signify especially decrease of consumption, which in fact has not occurred. Instead, consumption in western industrialised societies has constantly been growing. According to UN’s Human Development Report global consumption has more than doubled in almost thirty years, public and private consumption expenditures reached $24 trillion in 1998 - twice the level of 1975 and six times that of 1950 (Hahn 2002, 41). And on average, resource use per person nearly tripled between 1950 and 1990 (Corson 1994). One explanation to the state of affairs is that the growth of industry and production in the last century yielded not only increased common wealth but it also produced a new kind of pleasure to consumers: delights of motoring, hanging out on malls and shopping (Vahvelainen 2001, 3). Therefore, it is not solely question about production and its’ influence to the nature but about consumption and the role of consumers in shaping environmentally induced social change (Spaargaren 1999). Consumer behaviours frequently take such behavioural forms that have malign influence to the environment. Environmental critique towards consumption and consumerism is mostly critique towards destructive consequences they involve (Gabriel – Lang 1995, 26, Foxall 1995, 263–264; Slater 1999). This is on way to see consumption’s impacts on the environment. The other way is to see consumption as a factor which promotes environmental improvements in society. Keeping in mind the facts about the influences of consumption but still recognising the dynamics of western industrial societies helps to understand the problematic of contemporary consumption. 57 Due to the problems related to the impacts of consumption towards the environment, more research is needed in the consumer behaviour field. This paper aims at giving contribution to ongoing debate concerning environment, consumption and consumer behaviour by monitoring micro level’s consumption, i.e. consumers’ role in environmental questions. These notions are approached from conceptual viewpoint by connecting sociological and environmental perspectives on consumer behaviour research. The conceptual framework is hence, offered by environmental sociology and especially ecological modernisation theory, which core lies in interactions between environment and society (Dunlap – Catton 1979; Sonnenfeld – Mol 2002b; Spaargaren 1999). 3.2 Green consumption, consumers and the environment: definitions Writing about environmental issues employs a wide range of metaphors and rhetoric sentences. Concepts are often replaced with metaphors of nature as it has been understood in western societies. (Harré, Brockmeier – Mühlhäusler 1999, 91–118.) “Less is more” (Peattie 1995) and “Small is beautiful” (Schumacher 1973, in Murphy 2001) are examples of often used slogans related to promotion of green products, as well as “looking world through green-coloured glasses” is just a way of greenspeak and in fact seeks to questioning the greening of society (Harré et al 1999, 97). It is therefore, not an easy task to try to determine the concepts related to environmental issues, but an attempt is done beneath. The notion of green is often used as a synonym for the term environmental (e.g. Peattie 1999, 133; Wagner 1997, 26). Green implies to underlying concern for preservation of the environment and a non-invasive lifestyle (Iyer – Banerjee 1993, 494). An essential question related to environmentally responsible or green consumption is, whether it is possible to define proper green consumer behaviour. Ideally, green consumer behaviour deals with such consumption manners that impact environment so little as possible. According to this view green consumer should assess the greenness of a product on the basis of life- cycle analysis (LCA) (Wagner 1999, 24–25). The theoretical aspect of LCA concept is problematic; because the evaluation and analysis of choices made by consumers in every day situation is difficult to indicate. Thus, it seems that the employ of terms green consumption and green consumers are to a great deal depending on the substance of the discussions. Writings concerning green consumers refer on the one hand to radical and marginal groups, deep ecologists or environmental movements (cf. Konttinen 58 1999, 46). On the other hand, green consumers are seen mostly as “normal citizens”, which have adopted greenness in some level of action as part of everyday life (Wagner 1997, 25–26, Autio – Wilska 2003, 4–5). The greenness of consumer behaviour is contextual in nature connected with skilled behaviour in everyday life. Being and acting green is characterised in the literature by daily consumption trade-offs related to second hand shopping, product choices and other behaviours enhanced by environmental attitudes, values and motives and can even be considered as a holistic consumption systems (cf. Pantzar – Heiskanen 1995, Laaksonen – Mäntylä 2000). The term green is typically used in free way and in loose context when expressing consumers and their behaviour and thus, single definition can be given neither to the green consumption or green consumer. Environment in studies within environmental sociology’s field is determined as factor that may influence or be influenced by human behaviour. Interactions can be divided into three levels: cognitive, behavioural, and physiological interactions between humans and environment (Dunlap – Catton 1979, 253). Dunlap and Catton (1979) suggest that all of these interaction levels are worth paying attention, because otherwise it would be difficult to understand the complexity of interactions between human and the physical environment. The researcher interested in environmental sociology and behavioural issues should understand all human interaction levels –cognitive, behavioural and physiological– and also their combinations to be able to consider all aspects of interactions (Dunlap – Catton 1979, 254). On the other hand Wagner (1997, 14) points out that due to the complexity of green consumer behaviour only limited elements of the phenomenon can be researched at a time. By studying these mutual interactions as Dunlap and Catton propose or restricted interactions as Wagner suggests one could seek answers to, how the different systems in the society influence the physical environment. Essential question is then, whether these systems are changing their ways of operation into a more environmentally responsible direction. And particularly, what is the role of micro-level consumption in encouraging these changes. The analytical framework of environmental sociology developed by Dunlap (in Dunlap – Catton 1979, 251) is based on the concept of the ‘ecological complex’. It focuses on the interdependence among population, organization, environment, and technology (P,O,E,T), which emphasises the reciprocity of each element, stressing the ‘E’ not as social environment but rather as physical environment. Basically, tasks of environmental sociology are to seek answers to two kinds of questions: Firstly, how do interdependent variations in population, technology, culture, social systems, and personality systems impact on the physical environment. Secondly, how do resultant changes (and 59 other variations) in the physical environment modify population, technology, culture, social systems, and personality systems, or any of the interrelations among them? (op.cit. 252.) The core of ecological modernisation theory as mentioned is studying social change, environmental improvements in social practices, institutional designs and societal and policy discourses (Mol – Sonnenfeld 2000, 5; Murphy 2001, 1). As developed by sociologists, this theory sees the environmental crisis stemming from social crisis, from the organisation of production and consumption in modern societies (Spaargaren 1999; Murphy 2001, 1). The environmental threats caused by contemporary western consumerism are therefore very much social crisis in nature. Green consumption and consumers in this paper are elaborated by looking them through environmental sociology and more specifically through the ecological modernisation theory. It is not in the main interest of this paper to examine environmental sociology or the theory of ecological modernisation as per se, but mainly to discuss, how these domains employ consumer behaviour. Therefore, the key characteristics of both domains are summed up in the following Table. I. Table seeks to sum up the treatment of concepts environment and consumption heading from general level to more specific distinguishing environmental sociology (ES) from the theory of ecological modernisation (EMT). Table 1. Characteristics of ‘environment’ and ‘consumption’ within sociology, environmental sociology and ecological modernisation (based on Baudrillard 1998; Corrigan 1997; Dunlapp – Catton 1979; Buttel 1987; Spaargaren 1999; Murphy 2001) Sociology Environmental sociology (ES) Ecological modernisation theory (EMT) Environment means cultural and social influences upon behaviour Environment in ES is a factor that may influence or be influenced by human societies and behaviour. Environment Environment in EMT is viewed in terms of energy and material flows through physical and social systems Consumption is an order of significations. It communicates positions in society. Consumption is seen mostly related to environmental behaviour, attitudes and motives. Consumption is seen as key concept to a better understanding of the dynamics of industrial societies 60 The characteristics of sociology, environmental sociology and ecological modernisation as a system of ideas about the society-environment relationship are perhaps best summarised by contrasting them with those of environment and consumption concepts as shown in the above Table 1. It must be held in mind that these characteristics are not exhaustive by any means, but aim to be more a basic summary. Ecological modernisation theory is an attempt to understand the impact on society of environmental problems that began to emerge in the late 1970s (Spaargaren 1999, Murphy 2001, 4).18 Central view to EMT is that modern innovations, such as technological development, can lead to greening of business. Within EM theory, environmental technology plays an important role (Spaargaren 1999). Ecological modernisation theory is not very radical neither very strict when it comes to assumptions of growth of production and consumption and their ecological impacts. This mildness has also caused major criticism against it (see e.g. York – Rosa 2003, 278–279. York and Rosa have also criticised EM theory when it comes to individual consumption analysis. They ask whether there is sense in applying EM theory to micro-, individual consumers level because this kind of evidence is not telling enough of sustainable lifestyle transformation (York – Rosa 2003). The following chapters try to give one response and view about this crucial point within the ecological modernisation theory. 3.3 Green consumption and social change Social change is a major theme in EMT. Social change approaches greening consumption as changing social constellation (Spaargaren 1999; Mol – Sonnenfeld 2000). Environment induced social change according to Spaargaren (1999) refers to the demand of a theory that would take into account broader social dynamics. Nations encounter changing interdependencies, for example, new coalitions and levels of governance that have to be embodied in the theory of EM (Sonnenfeld – Mol 2002b, 1457). There is not only one way to achieve social change but “many routes with different sets of social actors and involving different social mechanisms” (Spaargaren 1999). 18 In this paper is not given an overview of the ways in which EMT has been developed during the last twenty years. For more about historical phases of EMT see e.g. Spaargaren, G. & Mol, A. (1992) Sociology, environment and modernity: ecological modernisation as a theory of social change. Society and Natural Resource. Vol, 5, 323–344. 61 In studying social change Spaargaren also makes division between macro and micro level consumption. The division is important firstly, in avoiding the risk to loose the individual actor’s role in bringing about environmental transformation and secondly, when considered social practices in production- consumption-cycles (Spaargaren 1999). Sonnenfeld and Mol’s approach to social change is related to market-based environmental reform (Sonnenfeld – Mol 2002a). It refers to the growth of environmental consciousness that leads to consumer-oriented mechanisms. Greening consumerism manifests itself in green consumption and purchase choices. This orientation involves the interplay firstly, between economic actors (or agents) and market dynamics and secondly, between citizen- consumers and political institutions in environmental reform. (Sonnenfeld – Mol 2002a; Mol – Sonnenfeld 2000.) Key point in their view is that environmental authorities are no longer the sole responsible agency for designing, implementing, and solving environmental problems (Sonnenfeld – Mol 2002b, 1457). This view emphasizes consumers’ role in the way to more sustainable society. Contemporary western consumerism is often related to rampant materialism and mass consumption, where consumers are having never ending consumption parties, extravagance life styles and therefore littering the environment (Vahvelainen 2001, 3; Benstein 2001, 311). The environmental crisis appears when consumer meets his/her limits in an environment which cannot sustain the needs generated within modern consumer culture (Slater 1999). These kinds of views about the consumerism highlight the individual’s, human agency’s (cf. Spaargaren 1999) role and responsibility in environmental issues. And yet, it is not only a matter of responsibility, but acknowledging the increasing importance of market dynamics and economic agents, in which customers and consumers are significant factors. Paying attention to the role of consumption in general and its various aspects helps to understand green consumption more deeply. Traditionally, occupation and work shaped people’s considers about themselves, to date, people seek their identity mostly by buying and consuming goods (Murphy 2001, 11). Things are purchased not for, what they are, but for what they mean. The most characteristic feature “a mystery” of modern consumerism concerns “the very essence of modern consumption itself – its character as an activity which involves an apparently endless pursuit of wants” (Campbell 1987, 37). In sociology, consumption has at least two aspects: identity formation and group communication between members (Burgess 2003, 80, 230, Murphy 2001, 11–12). 62 Identity formation is closely linked to all consumption. Western consumers can be presented as “thirsting for identity” when evaluating, choosing and buying goods (Gabriel – Lang 1995, 81). Also linkages between consumption and lifestyle are closely related to the seeking of identity. Products are building linkages to lifestyles by constructing a bridge to the ideal lifestyle (McCracken 1988, 113 in Corrigan 1997, 46–47). Furthermore, consumers use the set of products, consumption constellations, to define, communicate and perform social roles (Solomon, Bamossy – Askegaard 1999, 405). This is very typical in designing products and brands, which are personalised by adding personal features of individuals to them (cf. Hakala 2003, 9–10). It is useful to consider separately, as Murphy (2001, 12) suggests group communication from identity formation. Such concepts as perceptions of taste and status buying and status symbols express people’s tendency to evaluate themselves relative to others (Solomon et al 1999, 346–347). The Veblenian term “conspicuous consumption” refers to this very role of consuming goods in order to inspire envy in others (Solomon et al. 1999, 347; Corrigan 1997, 21–26). Corrigan draws interesting analogies of Veblen’s concept of “feudalism” to “future ‘green’ society?” (Corrigan 1997, 26). Can conspicuous leisure and feudalist society as its manifestation be seen as future green society? Corrigan doesn’t give any precise explanation to this proposition. It could be understood anyhow, that with the green society he means that in feudalist society “where everyone knows everyone else” people have far less people to impress and therefore less need to consume (op.cit. 25–26). 3.4 Green consumption and micro-level consumers Conventionally natural environment has been something external for the economic activity (cf. Slater 1999). For the most part, in economic models and in particular in consumer behaviour models, people and nature has been posed against each other (Massa 1995, 261; Slater 1999). Consumption and role of consumers has gathered quite little attention so far in EMT. In this subchapter the aim is to put together some central themes regarding the micro-level consumption as inducing the social change. Research focusing on environmental behaviour is one of major areas within environmental sociology (Buttel 1987, 472–475). Topics such as littering, segmenting through social class and environmental concern are typical orientations (Buttel 1987, 472, van Liere – Dunlap 1980). Nonetheless, study 63 of environmental consumer behaviour seems to be scarce.19 Wagner 1(997, 22) and Olney and Bryce (1991) in their reviews noted that in consumer research sociologically (or social marketing) qualified studies are scarce and even more rare are studies dealing with green consumer behaviour from sociological perspective. An example of green consumer behaviour research that touched upon sociological issues is Olney and Bryce (1991) research of customer’s confidence in unconvincing ‘environmentally sound’ products and companies (see also Wagner 1997, 22). One of the environmental sociologists, Frederick Buttel (1987) has given particular attention to environmental behaviour (although not approaching it from consumer view). His contribution to environmental sociology basis not only to his wide analysis of Catton – Dunlap’s and Schnaiberg’s works but also to critical theoretic and complementary review of the development of environmental sociology in the 1980s. Buttel focuses in his 1987 paper to major clusters of environmental sociology during 70s and 80s. Particular attention in this work is given to research on environmental attitudes and values. There can be distinguished three kinds of literature on attitude, value, and behaviour research: studies examining social-structural problematic, studies examining social-psychological theory, applied studies attempting to determine the social factors related to behaviour associated with the environment (Buttel 1987, 472). Social-structural aspects of environmental attitudes are based on surveys dealing much with public concern about the quality of the environment. Main correlates are socio-economic variables. Strongly related variables were education and especially age, which was negatively correlated with environmental concern. Even better explainers of environmental concern have been indicators of political ideology. Liberalism was positively correlating with environmental concern, suggesting that the alleged elitism –that environmental concern would be only an elitist issue– was not supported in the studies reviewed by Buttel. (1987, 473–474; see also van Liere – Dunlap 1980, 189.) Social-psychological research involves theory testing and evaluation of studies made in field. The target is on the studies of cognitive structure of environmental behaviour and attitude-behaviour congruence. Heberlein (1981, in Buttel 1987, 474) points out that in environmental concern it is question about people’s opinions rather than attitudes and because opinions are not stable cognitive structures they are not likely to affect behaviour significantly. 19 Conversely, in consumer behaviour research, articles dealing with social issues are overall few as reported by Richins (2001, 2). According to her, studies including social variables featuring prominently comprised only 10 % of the total of articles published in Journal of Consumer Research on a twelve-year time scale. 64 Appealing in social-psychological research area is that environmental behaviours or attitudes toward environmental protection reflect social- psychological processes involving the action of moral norms against harming people (Stern et al 1986, in Buttel 1987, 475). The theory of moral norms seems rather interesting regarding also when thinking about every-day shopping behaviour. The content of moral norms theory is about actions and consequences: if consumers are aware of what kind of negative consequences their actions cause, their moral norms will be activated and they are expected to accept personal responsibility for the consequences of these actions (ibid.). Given this, it is possible to illustrate that consumers when choosing among different products, would –if activated as presented according to the moral norms theory– prefer environmentally sound products instead of ordinary ones. Inspired by this it could be hypothesised that consumers in open setting, in such purchase situation where the amount of control is minimum and choice options are extensive (Foxall 1995), will choose the environmentally sound product instead of an ordinary product, because they become aware of the negative consequences of harmful goods to the environment. The last of Buttel’s (1987, 475) review on environmental attitudes and behaviour concerns applied research. The studies touched upon social demarketing (see also Foxall 1995), intentions to influence environmentally related behaviours by means of incentives, information etc. Buttel concluded that: “while economic motivations and incentives may have some impact, these factors tend to be less important than nonfinancial motives, effective communication and information, and the trustworthiness of information and sponsoring organizations” (1987, 475). Van Liere and Dunlap (1980) reviewed in their article environmental concern on the grounds of existing knowledge regarding the social bases. The social and demographic variables were age, social class, residence, political ideology, and sex. From the literature that they went through (21 studies) they generally found evidence for the hypothesis, but the success was moderately limited. In general the coefficients were quite low, even in the cases of age, education, and political ideology (Van Liere – Dunlap 1980, 193). As a result, predicting concern of environment’s quality with social variables is not sufficient, but rather one influencing factor. Spaargaren (1999; Spaargaren 1997, in Murphy 2001) of the scholars of ecological modernisation theory has raised the consumer-led perspective as a crucial issue for the development of the theory. What distances the ecological modernisation of consumption from many other proposals offered by the environmentalists and technologists is that in the EM theory technology has an important part to play. This suggestion to the industrialised society involves however, acknowledging social change and 65 processes that underlie behavioural changes in consumption. (Murphy 2001, 13; Spaargaren 1997, 169 in Murphy 2001, 13) Environmentally friendly consumption is not an objective as such. The goal of consumption in EMT is rather to achieve such consumption practices that consumers themselves consider as environmentally friendly (Spaargaren 1997, in Murphy 2001, 13.) Especially and due to the sociological nature of EMT, evaluation of micro- level consumption is important in order not to loose the significance of human actor. This comes into question particularly in holistic theoretical modelling (Spaargaren 1999.) Micro-level research, i.e. exploring consumers’ role in consumption practices is thus, essential. As discussed already before in this paper, consumption is not just satisfying basic needs but is much individualistic in nature (Baudrillard 1998, 5, see also Slater 1999) and symbolizes distinctive lifestyles (Burgess 2003, 80). Spaargaren and Van Vliet (in Burgess 2003, 81, see also Spaargaren 1999) have developed a conceptual model which helps to understand the various interconnection of consumption practices more clearly. Actors-------------Human Action-------------Social Practices---------Structures Discursive & Practical Conscious- ness L I F E S T Y L E Practice 1 Practice 2 Practice 3 Etc. Rules & Resources S Y S T E M of P R O V I S I O N Figure 1. A conceptual model for studying consumption practices (Spargaaren – Van Vliet 2000, in Burgess 2003, 81). 66 The above Figure illustrates how the home-bound practices (e.g. heating, bathing, and cooking) are not isolated functions but merely closely connected to society as whole. These bundles of practices make up a lifestyle. What we eat or what leisure activities we enjoy function as sign of our individual values. Spargaaren (1999) underlines the importance of home-based social functions for the study field in environmental sciences where “our relation with the sustenance base, our daily interaction with nature on a routine base, is shaped.” (1999, emphasis in the original). As the figure illustrates in the left-hand side of the diagram, these practices are guided by discursive and practical awareness. There is not much sense in investigating consumption from a home-bound perspective only but looking consumption and consumerism in larger scale, consumers in different roles (cf. Gabriel – Lang 1995). System of provision as well as rules and resources on the right-hand side of the diagram identifies this interdependency. Laws and institutional orders on one hand, and socially acceptable norms and codes on the other hand conduct human behaviour. 3.4.1 Future perspectives on consumer studies within ecological modernisation theory The growing prominence of environmental issues and the increase in environmental awareness has already caused changes throughout society and these changes are likely to continue in the future. While the rapid continuation of growth of economy on environments cost (cf. exploitative economy in Massa 1995) goes on, research in academic field is needed. There has been criticism against the ecological modernisation theory (Rosa & York 2003). Solid theoretical bases for accepting the consumer’s role in relationship of micro-level consumption and environment would certainly diminish such disagreement. Empirical studies concerning different aspects of micro-level consumption, consumer behaviour are needed. Consumers' acts and behaviour have no doubt impact on environment state. But as the global consumption is increasing the criticism is not the way to solve the problems. In contrary, once it is understood the importance of consumers’ role in suggesting social change and in improving the environments’ state – then it will be easier to see the dynamics of industrial societies both in macro- and micro-level. 67 References Autio, Minna – Wilska, Terhi-Anna (2003) Vihertävät tytöt ja vastuuttomat pojat: Nuorten kuluttajien ympäristöasenteet. Nuorisotutkimus, Vol. 21, No. 2, 3–18. Baudrillard, Jean (1998) The consumer society: Myths & structures. SAGE Publications: London. Benstein, Jeremy (2001) Alma de’atei. The-world-that-is-coming: Reflections on power, knowledge, wisdom, and progress. Cross Currents, Vol. 51, No. 3, 308–321. Burgess, Jacquie (2003) Sustainable consumption: Is it really achievable? Consumer Policy Review. Vol., 13, No. 3, 78–84. Buttel, Frederick H. (1987) New directions in environmental sociology. Annual review of Sociology, Vol. 13, 465–488. Campbell, Colin (1987) The romantic ethic and the spirit of modern consumerism. Basil Blacwell Ltd: Oxford. Corson, W. H. (1994) Changing the course: An outline of strategies for a sustainable future. Futures: The Journal of Forecasting and Planning. Vol. 26, No. 2, 206–223. Dunlap, Riley E. – Catton, William, R. Jr. (1979) Environmental sociology. Annual Review of Sociology, Vol. 5, 243–273. Dunlap, Riley, E. – Catton, William, R. Jr. (1994) Struggling with human exemptionalism: The rise, decline, and revitalization of environmental sociology. The American Sociologist, Vol. 25, No. 1, 5–30. Foxall, Gordon R. (1995) Environmental-impacting consumer behaviour: An operant analysis. Advances in Consumer Research, Vol. 22, No. 1, 262–268. Gabriel, Yiannis – Lang, Tim (1995) The unmanageable consumer: Contemporary consumption and its fragmentations. SAGE Publications Ltd: London. Hahn, Tamar (2002) The missing link. Tomorrow: Global Sustainable Business, Vol. XII, No. 2, 40–42. Hakala, Ulla (2003) Quantitative and qualitative methods of analysing advertising: Content analysis and semiotics. Publications of the Turku School of Economics and Business Administration. Series discussion and working papers 5:2003: Turku. Harré, Rom - Brockmeier, Jens – Mühlhäusler, Peter (1999) Greenspeak: A study of environmental discourse. SAGE Publications, Inc: Thousand Oaks, CA. 68 Järvikoski, Timo (1996) The relation of nature and society in Marx and Durkheim. Acta Sociologica, Vol. 39, No. 1, 73–86. Konttinen, Esa (1999) Four waves of environmental protest. In: All Shades of Green: The Environmentalization of Finnish Society, ed. by Esa Konttinen, Tapio Litmanen, Matti Nieminen and Marja Ylönen, 20–46. University of Jyväskylä, SoPhi: Jyväskylä. Laaksonen, Pirjo – Mantylä, Miia (2000) Tuotteen ympäristöystävällisyys: Kuluttajien ja asiantuntijoiden käsitysten vertailua. Kuluttajatutkimuskeskus, Julkaisuja 2/2000. Massa, Ilmo (1995) Historical approach to environmental sociology. The European Journal of Social Sciences, Vol. 8, No. 3, 261–274. Mol, Arthur P. J. – Sonnefeld, David A. (2000) Ecological Modernization around the world: an introduction. Environmental Politics, Vol. 9, No. 1, 3–16. Murphy, Joseph (2001) Ecological modernisation: The environment and the transformation of society. OCEES Research Paper No 20. , 21.10.2003. Olney, T.J. - Bryce, Wendy (1991) Consumer responses to environmentally based product claims. Advances in Consumer Research, Vol. 18, No. 1, 693–696. Pantzar, Mika – Heiskanen, Eva (1995) Toward sustainable consumption: Two perspectives. Consumer Research Center, Työselosteita ja esitelmiä 16/1995. Peattie, Ken (1995) Environmental marketing management: Meeting the green challenge. London: Financial Times Management. Peattie, Ken (1999) Trapping versus substance in the greening of marketing planning. Journal of Strategic Marketing, Vol. 7, No. 2, 131–148. Slater, Don (1999) Themes from the sociology of consumption. Solomon, Michael – Bamossy, Gary – Askegaard, Søren (1999) Consumer behaviour: A European perspective. Prentice Hall Inc: New Jersey. Sonnenfeld, David A. – Mol, Arthur P. J. (2002a) Globalization and the transformation of environmental governance. American Behavioral Scientist, Vol. 45, No. 9, 1318–1339. Sonnenfeld, David A. – Mol, Arthur P. J. (2002b) Ecological modernization, governance, and globalization. American Behavioral Scientist, Vol. 45, No. 9. SAGE Publications, 1456– 1461. Spaargaren, Gert (1999) The ecological modernisation of domestic consumption. 69 , 21.10.2003 Wagner, Sigmund A. (1997) Understanding green consumer behaviour: A qualitative cognitive approach. London: Routledge. Vahvelainen, Simo (2001) Ympäristöyhteiskunnan näköinen paikka. Hyvinvointikatsaus 2/2001, 2–6. van Liere, Kent, D. – Dunlap, Riley E. (1980) The social bases of environmental concern: A review of hypotheses, explanations and empirical evidence. Public Opinion Quarterly, Vol. 44, No. 2, 181–197. York, Richard – Rosa, Eugene, A. (2003) Key challenges to ecological modernization theory: Institutional efficacy, case study evidence, units of analysis, and the pace of eco-efficiency. Organization & Environment. Vol., 16, No., 3, 273–288. 70 71 4 MATKAILUN KAUSILUONTEISUUS Antti Honkanen Tiivistelmä Artikkelissa tarkastellaan matkailun kausiluonteisuutta, jota pidetään eräänä matkailuelinkeinon pahimpana ongelmana, kahdessatoista Euroopan unionin jäsenvaltiossa vuosina 1985 ja 1997. Artikkelin alussa keskustellaan, mitä matkailun kausiluonteisuudella tarkoitetaan ja mitkä tekijät aiheuttavat sitä. Tämän jälkeen esitellään kausiluonteisuuden mittaamiseen tarkoitettuja deskriptiivisiä menetelmiä. Lopuksi tutkitaan kausiluonteisuuden voimakkuutta kahdessatoista Euroopan unionin jäsenvaltiossa vuosina 1985 ja 1997. Tulosten mukaan uusia matkailukausia ei juuri ole syntynyt ja kausiluonteisuus on säilynyt yllättävän samankaltaisena. Maiden väliset erot ovat kuitenkin kohtuullisen suuret. Kahden käytetyn menetelmän, kausiluonteisuusindikaattorin ja gini-kertoimen avulla saadut tulokset ovat samankaltaisia, joskin myös eroja on löydettävissä. Myös graafisen kuvailun merkitys korostuu. ASIASANAT: matkailun kausiluonteisuus, kausiluonteisuuden mittaaminen, matkailun muutokset 4.1 Kausiluonteisuus matkailun ongelmana Matkailun kausiluonteisuutta on usein pidetty eräänä matkailuelinkeinon pahimpana vitsauksena, joskin siitä on löydettävissä myös eräitä positiivisia puolia kuten se, että matalasesonki auttaa matkakohteen väestöä ja luontoa toipumaan huippusesongista (Toivonen 1999). Sekä matkailuelinkeino että – tutkimus on kiinnittänyt paljon huomiota sen torjumiseen. Kausiluonteisuus ei ole pelkästään matkailussa esiintyvä ilmiö, vaan sitä esiintyy lähes kaikkien tuotteiden myynnissä. Matkailussa, kuten monissa muissakin aineettomissa tuotteissa tai palveluissa, kausiluonteisuuteen liittyvä ongelma korostuu: matkailutuotteen varastointi huippusesonkia varten on vähintäänkin ongelmallista. 72 Matkailun kausiluonteisuudella tarkoitetaan matkailijavirtojen epätasaista jakautumista vuoden eri kuukausina (Butler 1994, 332), joskin ilmiö on havaittavissa muillakin tasoilla, kuten viikon eri päivinä. Matkailun sesongit on usein jaettu huippu– (peak), väli– (shoulder) ja matalasesonkiin (off-peak), joka jossain tapauksissa saattaa olla myös ei-sesonki – aika jolloin matkailijoita ei ole ollenkaan. Sesongeille ei ole mitään tarkkaa määritelmää, vaan sesonkien nimeämisessä valinnan tekee useimmiten tutkija (Honkanen 2000, 6) tai ääritapauksessa kyseessä voi olla jopa poliittinen kysymys (ks. Kennedy & Deegan 2001, 52). Esteban Bardolet ehdottaa, että sesonkiajasta on kyse silloin, kun yöpymisten määrä kuukaudessa on yli puolet parhaimmasta kuukaudesta (Bardolet 1999). Toisaalta esimerkiksi tässä käytetyssä aineistossa tämä olisi johtanut siihen, että vain heinä- ja elokuu olisivat täyttäneet sesongin määritelmän. Tällöin käyttökelpoisempi ratkaisu olisikin, että sesongit määritellään mahdollisimman yksinkertaisella tavalla esimerkiksi dikotomiaksi niin, että matkailusesonkia ovat ne kuukaudet, jolloin matkalle lähteneiden määrä on suurempi kuin se olisi, mikäli matkat olisivat jakaantuneet tasaisesti jokaiselle kuukaudelle. Vastaavasti pienemmän arvon saaneet kuukaudet olisivat tällöin sesongin ulkopuolista aikaa. Kausiluonteisuuden aiheuttamien ongelmien lista on pitkä. Sen seurauksena matkailua varten rakennettu infrastruktuuri on osan vuodesta vajaa- ja toisen osan ylikäytössä, minkä lisäksi monet matkailun synnyttämät työpaikat ovat osa-aikaisia (Honkanen 2000, 6-7). Pahimmillaan kausiluonteisuus voi aiheuttaa korkean työttömyyden matalasesongin aikana ja työvoimapulan huippusesongin aikana (Lundtorp, Rassing & Wanhill 1999, 50). Nostaessaan infrastruktuurin käytön äärirajoilleen kausiluonteisuus aiheuttaa saastumista ja lisää liikenneonnettomuuksien määriä. Myös matkailutuotteen laatu kärsii huippusesonkien aikana liian monen matkailijan käyttäessä samaa resurssia. (McEniff 1992, 68.) Matkailuyritysten toiminta puolestaan vaikeutuu osan kuluista kertyessä koko vuoden ajalta tulojen ollessa osavuotisia (Honkanen 2000, 6). Ylipäätään matkailun kausiluonteisuuden voidaan sanoa näkyvän kaikilla matkailuelinkeinon tasoilla (Baum 1999, 5). Matkailun kausiluonteisuuden aiheuttavia tekijöitä ovat muun muassa vuoden kiertoon liittyvät asiat, kuten sääolosuhteet, ja sosiaaliset käytännöt. Puhutaankin luonnon kiertokulusta johtuvasta luonnollisesta kausiluontoisuudesta ja ihmisen toiminnallaan ja käytännöillään aiheuttamasta institutionaalisesta kausiluonteisuudesta. Jälkimmäisellä viitataan muun muassa edellä mainittuun yhteisöjen tapaan järjestää työ- ja vapaa-ajan vaihtelu sekä juhlapäivät ja festivaalit. (Butler 1994; Frechtling 1996, 55.) Butlerin (1994) mukaan kausivaihtelua selittävinä tekijöinä voidaan pitää edellä mainittujen lisäksi sosiaalista painetta ja muotia, urheilusesonkeja sekä traditioita. Frechtling (1996) puolestaan mainitsee liike-elämän vakiintuneet 73 tavat ja kalenteripyhien (esim. pääsiäinen) sijoittumisen ja niistä seuraavat pitkät vapaat. Aiempien tutkimusten perusteella voidaan olettaa, että sosiaaliset käytännöt ovat eräs suurimmista kausiluonteisuutta selittävistä tekijöistä (esim. Honkanen 2000). 4.2 Sosiaalisten käytäntöjen merkitys Työn ja vapaa-ajan jaksottelua voidaan pitää suurimpana kausiluonteisuutta ylläpitävänä voimana moderneissa yhteiskunnissa. Vapaa-ajan matkustaminen on työssä käyville tai opiskelijoille mahdotonta ilman vapaapäiviä, jolloin loma-aikojen merkitys matkailun kausiluonteisuudelle on erittäin suuri. (Hartman 1986, 25–26.) Suosituin matkustuskuukausi onkin lähes poikkeuksetta kuukausi, jolloin ihmiset viettävät vuosilomansa, ja yleensä kaikissa maissa on varsin joustamaton vakiintunut vuosiloman viettämisajankohta. Mikäli matkailijoiden vapaa-aika on sidottu perinteisiin sesonkiaikoihin, ei markkinointi yksinkertaisesti tehoa tai jonkin kohteen vähentynyt kausiluonteisuus on vain pois joltain toiselta kohteelta. Ilman työ- ja vapaa-ajan lisääntyvää joustoa kausiluonteisuuden poistaminen ei ole mahdollista. Viime vuosikymmeninä on yhteiskunnan rakenteiden kuitenkin uskottu kokeneen muutoksia, jotka vaikuttavat työn ja vapaa-ajan vuorotteluun (Rojek 2000), ja toisaalta esimerkiksi elämäntyylien moninaistumisen (Featherstone 1995) oletetaan vähentävän sosiaalista painetta elää samalla tavalla kuin kaikki muutkin. Tuotantorakenteiden muutos kohti joustavuutta (ks. Bell 1974; Miles 1998) voisi mahdollistaa lomat traditionaalisen kesälomasesongin sijasta myös muina aikoina. Koska joustavan tuotannon periaatteisiin kuuluu toimittaa tavara mahdollisimman nopeasti ilman varastointia, teollisuudella ei välttämättä ole edes mahdollisuuksia lopettaa toimintaansa kesälomien ajaksi, vaan heidän kannattaa rohkaista työntekijöitään pitämään lomiaan eri aikoihin tai vaihtoehtoisesti silloin kun kysyntä on matalimmillaan. Näin ollen tuotantorakenteen muutoksen voisi olettaa rohkaisevan ihmisiä tekemään lomamatkoja entistä useammin huippusesongin ulkopuolella. Postfordistisen tuotantotavan kausiluonteisuutta vähentävä vaikutus on kuitenkin rajallinen, sillä lapsiperheille koulujen lomilla on edelleen suuri vaikutus matkojen ajankohtaan. Toisaalta suurten ikäluokkien vanheneminen ja eläkkeelle siirtyminen oletettavasti lisäävät sesongin ulkopuolista matkailua. Ei ole mitään syytä olettaa suurten ikäluokkien matkustushalukkuuden loppuvan niin kauan kuin heillä riittää terveyttä, ja oikealla matkailutuotteiden hinnoittelulla on mahdollista saada heidän matkansa ajoittumaan sesonkien ulkopuolelle. Samankaltaista vaikutusta oletetaan olevan myös lyhytlomien 74 kasvavalla suosiolla (Edgar 2001, 71). Lyhytlomat eivät vaadi niin pitkää poissaoloa työpaikalta tai koulusta, jolloin niitä voidaan tehdä varsinaisten loma-aikojen ulkopuolella. Tutkimuksen tarkoituksena on esitellä, miten kausiluonteisuuden voimakkuutta voidaan mitata sekä poikkeaako kausiluonteisuuden voimakkuus maittain ja tarkasteltavina vuosina. Tutkimus ei varsinaisesti tarkastele kausiluonteisuuden syitä, joskin kausiluonteisuutta kuukausien pohjalta kuvaavista kuvioista voidaan vetää varovaisia johtopäätöksiä sosiaalisten käytäntöjen säilymisestä. Mikäli lomat keskittyvät myös jälkimmäisenä vuotena yhtä voimakkaasti pääasialliseen lomakauteen kuin vuonna 1985, voidaan olettaa sosiaalisten käytäntöjen säilyneen kausiluonteisuutta ylläpitävänä voimana. 4.3 Aineisto Tutkimusaineiston muodostavat Eurobarometrit 25,0 (Rabier, Riffault & Inglehart 1986) vuodelta 1985 ja 48,0 (Melich 1999) vuodelta 1997. Aineistoa voidaan pitää jossain määrin vanhentuneena, mutta on otettava huomioon, että mitään samankaltaista julkista aineistoa ei ole saatavilla ennen kuin matkailu otetaan seuraavan kerran Eurobarometrien aiheeksi. Eurobarometrien historia alkoi 1970-luvulla, jolloin Euroopan yhteisön toimesta alettiin tehdä tutkimussarjaa, joka keskittyi alussa lähinnä tietoisuuteen ja asenteisiin Euroopan yhdentymistä kohtaan. Myöhemmin kysymyssarjoja on laajennettu varsin monenlaisiin kysymyksiin, joista matkailua laajimmin käsittelevät juuri tutkimuksen aineistona olevat Eurobarometrit 25,0 ja 48,0. (Melich 2002.) Eurobarometrit on painotettu kummankin aineiston mukana tulevilla maakohtaisilla painoilla. Painotuksella vastaajat suhteutetaan maakohtaisesti NUTS 2 alueiden asukaslukuun huomioiden vähintään vastaajien sukupuoli, ikäluokka ja asuinalueen koko. Koko Euroopan unionia tarkastellessa vastaajat on suhteutettu Euroopan unionin asukaslukuun vastaavasti kuin edellä. (Standard Eurobarometer Weighting Overview 2003.) Kausiluonteisuutta tarkastellaan tässä sen mukaan, minä kuukautena haastateltavat ovat lähteneet matkalle. Analyysin kohteeksi on otettu vuoden pisin kodin ulkopuolinen loma ja kaikki tutkimuksissa ilmenneet vähintään neljän vuorokauden matkat. Vuoden päämatkat on luokiteltu myös sen mukaan, onko kyseessä kotimaan- vai ulkomaanmatka. Vuosien 1985 ja 1997 Eurobarometrit eivät ole täysin vertailukelpoisia keskenään: vuoden 1985 Eurobarometrissa loman alkamiskuukautta kysyttiin kahden pisimmän vähintään neljän vuorokauden matkan osalta. Sen sijaan vuonna 1997 kysyttiin korkeintaan kuutta haastateltavien satunnaisessa järjestyksessä mainitsemia 75 matkoja, joista osa on voinut olla alle neljän vuorokauden matkoja. Tässä alle neljän vuorokauden matkat on kuitenkin poistettu myös vuoden 1997 aineistosta. Kysymysten eroavaisuuksien vaikutusta tuloksiin vähentää kuitenkin, että useamman kuin kahden matkan tehneiden osuus on kohtuullisen pieni. Vuoden 1997 aineistossa ei myöskään ollut mahdollista yhdistää matkan alkamiskuukautta luotettavasti matkan kohdetta selvittävään kysymykseen, minkä vuoksi jaottelu kotimaan- ja ulkomaanmatkoihin on tehty pelkästään vuoden pisimmän matkan mukaan, jota tässä kutsutaan vuoden päämatkaksi. On kuitenkin syytä muistaa aineiston kattavan vain kaksi vuotta, jolloin varsinaista trendiä ei voida etsiä. Toisaalta aineiston merkitystä nostaa se, että sen avulla on mahdollista tarkastella matkalle lähtemistä yleensä käytetyn kohteeseen saapumisen sijasta. Yksittäisten kohteiden kausiluonteisuus voi vaihdella varsin monesta syistä, mutta kausiluonteisuuden merkitys koko matkailuelinkeinolle voidaan nähdä tarkastelemalla vain matkalle lähtemistä. Yksittäisessä kohteessa on markkinoinnin keinoin mahdollista saada matkailijoita entistä enemmän sesongin ulkopuolella, mutta ilman matkailijoita lähettävien alueiden loma-aikoihin vaikuttavien rakenteiden muutosta kyse on nollasummapelistä: sesongin ulkopuolella matkustamaan pystyvien matkailijoiden määrä on varsin rajallinen, jolloin jonkin alueen menestyminen sesongin ulkopuolella tietää toisen alueen kausiluonteisuuden nousua. 4.4 Kausiluonteisuuden mittaaminen Kausiluonteisuuden mittaamiseen on kehitetty varsin monenlaisia tapoja. Tilastotieteessä paljon käytettyä variaatiokerrointa (coefficient of variation) voidaan hyödyntää myös kausiluonteisuuden mittaamisessa (esim. Honkanen 2000; ks. Wanhill 1980; Yacoumis 1980). Variaatiokertoimen (CV) kaavassa s on keskihajonta ja x viittaa mitatun ajanjakson keskiarvoon: x sCV = Variaatiokerroin on varsin helppo laskea ja se on tuttu kaikille tilastotieteeseen tutustuneille. Lundtorpin (2001, 29) mukaan sen heikkoutena on kuitenkin, että sen avulla kausiluonteisuuden luonnetta on varsin vaikea tulkita – selittää mitä saadulla variaatiokertoimella tarkoitetaan – minkä lisäksi eri kuukausien välistä kausiluonteisuutta ei voida pitää stokastisena variaabelina. Yacomis (1980) käytti Sri Lankan kausiluonteisuutta käsittelevässä tutkimuksessaan kausiluonteisuuden suhdelukua R (seasonality ratio), jossa 76 verrataan suosituimman kuukauden matkustajamäärää ( ) koko vuoden kuukausikeskiarvoon ( nv v ) seuraavasti: vvR n /= Mikäli matkailijoita olisi joka kuukausi sama määrä, suhdeluku olisi 1. Mikäli kaikki matkailijat tulisivat suosituimman kuukauden aikana, arvo olisi 12. Kausiluonteisuuden suhdeluku on varsin selkeä indikaattori, mutta sitä muokkaamalla on mahdollista tehdä vielä paremmin ymmärrettävä suhdeluku. Lundtorpin (2001, 29) mukaan kyseisestä kausiluonteisuuden suhdeluvusta voidaankin ottaa myös käänteisluku, jolloin saadaan kausiluonteisuusindikaattori (seasonal indicator) ( ),ω joka voidaan laskea myös suoraan johtamalla edellä mainitusta kaavasta: nvv/ −=ω Kausiluonteisuusindikaattorin käyttäminen suhdeluvun sijasta on suositeltavaa juuri sen ymmärrettävyyden takia. Jos oletamme esimerkiksi, että jonkin hotellin suosituimman kuukauden käyttöaste on täydet sata prosenttia, voidaan kyseisen hotellin kausiluonteisuusindikaattori ymmärtää vuotuisena keskimääräisenä käyttöasteena. Tällöin kausiluonteisuusindikaattorin )(ω arvo 0,50 merkitsisi, että keskimäärin puolet hotellin huoneista olisi ollut käytössä vuoden aikana. (Lundtorp 2001, 30). Konkreettisen merkityksen johdosta kausiluonteisuusindikaattori on erityisen käyttökelpoinen silloin, kun kausiluonteisuutta kuvataan menetelmiä tuntemattomille. Tässä tutkimuksessa aineistona ei kuitenkaan ole tiettyyn tilaan mahtuvien matkailijoiden joukko, eivätkä kaikki haastateltavat ole matkustaneet, minkä vuoksi kyseinen tulkintaa ei ole mahdollinen, vaan kausiluonteisuusindikaattori ilmaisee vuotuisen matkustusasteen suhteessa suosituimpaan kuukauteen. Sekä kausiluonteisuuden suhdeluvun että kausiluonteisuusindikaattorin heikkoutena voidaan pitää, että ne eivät huomioi jakauman vinoutta ja että ne ovat herkkiä ääriarvoille. Onkin mahdollista, että jokin tilaisuus tai festivaali nostaa suosituimman ajanjakson kävijämäärän niin korkealle tasolle, että kausiluonteisuuden suhdeluku tai kausiluonteisuusindikaattori osoittaa korkeaa kausiluonteisuutta, vaikka muut mitattavat ajanjaksot olisivat kohtuullisen tasaisesti jakautuneita. Kausiluonteisuuden poistaminen ei voikaan olla itsetarkoitus, mikäli sen syynä on esimerkiksi jokin erityisen suosittu tapahtuma, joka aiheuttaa yksittäisen piikin matkailijamääriin. Näissä tapauksissa kausiluonteisuuden suhdeluku tai kausiluonteisuusindikaattori ei ole paras mahdollinen valinta kausiluonteisuuden mittaamiseen. Stephen Wanhill (1980) suosittelee näiden ongelmien vuoksi kausiluonteisuuden mittaamiseen taloustieteissä paljon käytettyä gini-kerrointa, jota voidaan pitää yleisimpänä tulojen epätasaisuutta mittaavana tunnuslukuna (Giles 2002, 1). 77 0,00 0,25 0,50 0,75 1,00 0,00 0,25 0,50 0,75 1,00 Lorenz plot for A Proportion of sample P ro po rti on o f v ar ia bl e Kuvio 1. Esimerkki Lorenzin- ja tasaisen jakauman käyrästä Gini-kerroin (G) perustuu Lorenzin-käyrään (ks. kuvio 1), jossa jakauman eri arvot asetetaan kumulatiivisesti nousevaan järjestykseen. Esimerkiksi matkojen jakautumisen epätasaisuutta vuoden eri kuukausina mitattaessa käyrän vasemmanpuoleisimmaksi arvoksi laitetaan vähiten suositun ajanjakson matkojen osuus koko vuoden matkoista. Seuraavaksi arvoksi tulee kahden vähiten suositun ajanjakson kumulatiivinen osuus kaikista matkustaneista. Kumulatiivista arvoa kasvatetaan aina seuraavaksi vähiten suositun kuukauden arvoilla, kunnes kaikki ajanjaksot on huomioitu. Tällöin saatua käyrää verrataan tasaisen jakauman käyrään. Gini-kerroin on suhdeluku, joka mittaa Lorenzin-käyrän ja tasaisen jakauman käyrän välisen alueen suuruutta. Mikäli Lorenzin-käyrä ja tasaisen jakauman käyrä ovat yhtenevät, gini-kertoimen arvo on 0. Suurin mahdollinen gini-kertoimen arvo on yksi. (Ks. Lundtorp 2001, 30–32; Moran 2003, 354– 356.) Gini-kerroin voidaan laskea seuraavalla kaavalla, jossa on fraktiilien (esim. kuukausien) lukumäärä, n iχ on fraktiilin järjestys (esim. 1/12, 2/12 jne.) ja iγ on fraktiilien kumulatiivinen arvo (Lundtorp 2001, 31): G= ∑ = − n i ii yxn 1 )(2 Gini-kertoimen ongelmana on kuitenkin, että se mittaa pelkästään Lorenzin- ja tasaisen jakauman käyrän välistä aluetta eikä näin ollen kerro mitään jakauman muodosta. Kahdella erimuotoisella jakaumalla voidaankin saada sama gini-kerroin. (Moran 2003, 357.) Sama ongelma koskee kuitenkin kaikkia edellä mainittuja kausiluonteisuuden indikaattoreita. Matkailun kausiluonteisuutta vertailtaessa pelkkää gini-kerrointa tai 78 kausiluonteisuusindikaattoria käytettäessä vaarana onkin, että varsin suuretkin kausiluonteisuuden muutokset jäävät paljastumatta. Ne kertovat matkailukausien epätasaisesta käytöstä, mutta ne eivät kerro mitään yksittäisistä kuukausista. Kausiluonteisuutta tutkittaessa olisikin kyseisten mittareiden lisäksi syytä piirtää esimerkiksi kuvio, josta on havaittavissa myös yksittäisten kuukausien muutokset kausiluonteisuuden suhteen. (Lundtorp 2001, 44–47.) Toisaalta usean vuoden sarjoista voidaan edellä esitellyistä kausiluonteisuuden numeerisista mittareista käyttää esimerkiksi variaatiokerrointa niin, että se lasketaan yksittäisen vuoden sijasta useiden vuosien eri kuukausien osalta, jolloin tuloksia voidaan tarkastella vuosien sijasta kuukausittain (ks. Koenig & Bishoff 2003, 238–239). Toisaalta vain muutamaa vuotta tarkasteltaessa kaikkein selkein kausiluonteisuuden luonnetta kuvaava mittari on usein matkustuskuukausi- indeksi. Matkustuskuukausi-indeksin käyttöä voi suositella koska, että sen avulla kausiluonteisuutta on mahdollista tarkastella havainnollisesti piirtämällä käyrästö matkustuskuukausi-indeksin pohjalta. Matkustuskuukausi-indeksin voi laskea seuraavalla kaavalla, jossa Tm on matkustaneiden määrä yksittäisen periodin (kuukausi) aikana, Ty matkailijoiden yhteenlaskettu määrä eri periodeissa (vuosi) ja Pn on periodien lukumäärä (12 kuukautta): SI= 100×     n y m P T T Matkustuskuukausi-indeksissä kaikki kuukaudet saisivat arvon sata, mikäli matkat olisivat jakautuneet tasaisesti ympäri vuoden. Yli sadan oleva indeksi kertoo kuukauden olevan tasaisen jakauman mallia suositumpi, jolloin ne voidaan katsoa sesonkiajaksi ja alle sadan olevan indeksin kuukaudet puolestaan sesongin ulkopuoliseksi ajaksi. Seuraavaksi tarkastellaan matkailun kausiluonteisuutta matkailijoita lähettävien maiden kannalta. Tätä varten on laskettu kaikkien Eurobarometreissä 25,0 ja 48,0 haastateltujen henkilöiden ilmoittamien kodin ulkopuolisten lomien ja vuoden päämatkojen alkamisajankohtien jakautumien perusteella kausiluonteisuusindikaattori ( ),ω gini-kerroin (G) ja matkustuskuukausi-indeksit (SI) vuosina 1985 ja 1997. Matkustuskuukausi- indeksin käyttöön on päädytty, koska kausiluonteisuutta on haluttu tarkastella graafisesti niin, että sitä voidaan verrata eri matkustustyypeissä. 79 4.5 Kausiluonteisuus Euroopan unionissa vuosina 1985 ja 1997 Kausiluonteisuusindikaattorien )(ω ja gini-kertoimien (G) arvojen välinen korrelaatiokerroin on -0,8220. Ne vastaavat kohtuullisen hyvin toisiaan, joskin myös eroja löytyy. Esimerkiksi vuoden 1985 kotimaan päämatkoissa Tanskalla oli kausiluonteisuusindikaattorin mukaan kaikkein suurin kausiluonteisuus, kun taas gini-kerrointa käytettäessä se oli vasta seitsemäs. Kausiluonteisuuden muutoksia tarkasteltaessa indikaattoreiden osoittama suunta on yleensä sama, josta poikkeuksen muodostaa samainen Tanska kotimaan päämatkojen suhteen. Siinä kausiluonteisuusindikaattori osoittaa lievää kausiluonteisuuden tasoittumista gini-kertoimen kertoessa kausiluonteisuuden lisääntymisestä. Matkustusindeksistä piirrettyä kuviota tarkastelemalla havaitaan, että heinäkuun, suosituimman lomakuukauden suosio kotimaan päämatkojen suhteen on lievästi pienentynyt, mikä selittää kausiluonteisuusindikaattorin kasvun. Kausiluonteisuusindikaattori on herkkä suosituimman kuukauden arvon vaihtelulle. Kausiluonteisuusindikaattorin käyttö vihjanneekin vähintään implisiittisesti, että tutkimuksen kiinnostus suuntautuu kausiluonteisuuden muutokseen suhteessa kaikkein suosituimpaan kuukauteen tai kausiluonteisuutta mitataan tilassa, jossa on jokin matkailijoiden maksimimäärä, johon pyritään vuoden jokaisena kuukautena. Kausiluonteisuuden muoto on säilynyt varsin samankaltaisena. Kaikkia Euroopan unionin (EU-12) jäsenvaltioita tarkasteltaessa kodin ulkopuolella vietettyjen lomien varsinaista sesonkiaikaa ovat olleet vain neljä kuukautta molempina vuosina. Suurin osa matkoista on tehty kesä-, heinä-, elo- ja syyskuussa. Myös gini-kerrointa tarkastelemalla havaitaan kausiluonteisuuden säilyneen lähes ennallaan, joskin kuviosta havaitaan elokuun piikin tasoittuneen jonkin verran. Koska kyseessä on vain kaksi vuotta, voimakkaita johtopäätöksiä tästä ei voida vetää. Eri maiden väliset erot ovat varsin suuret. Esimerkiksi Välimerenmaissa Ranskaa lukuun ottamatta on muita maita korkeampi kausiluonteisuus kaikkien matkojen kohdalla. Saattaa olla, että kyseisissä valtioissa on muita maita tiukemmat käytännöt vuosiloman pitämisajankohdan suhteen. Ulkomaanmatkojen suhteen näiden maiden kohdalla ei kuitenkaan voida vetää tarkkoja johtopäätöksiä, sillä matkustaneiden määrät olivat varsin vähäiset. 20 80,330, df=1, sig.= 0.000 80 Taulukko 1. Kausiluonteisuusindikaattorin ja gini-kertoimen arvot kaikilla matkoilla sekä päämatkoille kotimaassa ja ulkomailla Kaikki Matkat Päämatka kotimaahan Päämatka ulkomaille ω G ω G ω G 1985 1997 1985 1997 1985 1997 1985 1997 1985 1997 1985 1997 Kaikki 0,29 0,27 0,48 0,47 0,23 0,18 0,63 0,68 0,32 0,31 0,50 0,50 Iso- Britannia 0,39 0,41 0,44 0,38 0,31 0,25 0,54 0,58 0,37 0,44 0,43 0,40 Tanska 0,20 0,27 0,52 0,40 0,13 0,16 0,64 0,71 0,19 0,24 0,54 0,46 Saksa 0,38 0,41 0,43 0,41 0,27 0,24 0,52 0,58 0,30 0,38 0,54 0,50 Kreikka 0,27 0,19 0,59 0,67 0,25 0,16 0,66 0,72 0,34 0,26 0,40 0,44 Espanja 0,22 0,19 0,57 0,59 0,19 0,16 0,66 0,68 0,20 0,16 0,46 0,53 Ranska 0,27 0,26 0,47 0,50 0,21 0,19 0,64 0,71 0,31 0,24 0,48 0,57 Irlanti 0,27 0,35 0,52 0,50 0,19 0,20 0,66 0,64 0,36 0,36 0,49 0,48 Italia 0,20 0,17 0,62 0,66 0,17 0,14 0,71 0,76 0,18 0,16 0,63 0,64 Luxemburg 0,34 0,28 0,44 0,44 0,27 0,20 0,53 0,59 Alankomaat 0,23 0,32 0,50 0,40 0,19 0,20 0,71 0,68 0,19 0,25 0,65 0,53 Portugali 0,20 0,16 0,66 0,68 0,17 0,15 0,72 0,69 0,32 0,14 0,35 0,65 Belgia 0,24 0,25 0,59 0,53 0,22 0,17 0,61 0,60 0,20 0,22 0,59 0,60 Yleisesti ottaen kotimaassa tehtävissä päämatkoissa kausiluonteisuus on korkeampaa kuin ulkomaanmatkoissa. Poikkeuksia löytyy kuitenkin myös tästä, ja kausiluonteisuusindikaattorilla mitattaessa Belgiassa vuonna 1985 ja Portugalissa vuonna 1997 oli suurempi kausiluonteisuus ulkomaanmatkoissa kuin kotimaanmatkoissa. Vastaavasti gini-kerrointa käytettäessä näyttää sama tilanne olleen Saksassa vuonna 1985. Luxemburgin kotimaanmatkojen kausiluonteisuutta ei ole laskettu ollenkaan, sillä vuonna 1997 kukaan vastanneista ei ollut tehnyt päämatkaa kotimaassa eikä heidän määränsä ollut vuonna 1985 kovinkaan korkea. Kotimaanmatkoissa kausiluonteisuus on ollut muita pienempää Isossa-Britanniassa ja Saksassa, joskin vuosia vertailemalla havaitaan sen kasvaneen hieman. Eräänä varsin kiinnostavana seikkana voidaan mainita, että kotimaassa tehtyjen päämatkojen huippusesonki on siirtynyt heinäkuusta elokuuhun Isossa-Britanniassa ja Saksassa sekä Irlannissa toisinpäin elokuusta heinäkuuhun. Ulkomaille suuntautuneiden päämatkojen suhteen vastaava siirtymä on ollut Irlannissa ja Kreikassa. Edellisessä suosituin ulkomaan päämatkan kuukausi on muuttunut heinäkuusta kesäkuuhun ja jälkimmäisessä heinäkuusta elokuuhun. Kreikassa, Espanjassa ja Portugalissa ovat kuitenkin niin harvat haastateltavat tehneet ulkomaanmatkoja, että jo yhden matkan alkamiskuukauden muutos muuttaa matkustuskuukausi-indeksiä jonkin verran. Joka tapauksessa suosituimman kuukauden muutokset osoittavat sen, että myös sosiaalisiin käytäntöihin perustuvat matkustustottumukset voivat muuttua (ks. myös Syrjämää 2001), joskaan suuria muutoksia tarkasteltavalla aikajänteellä ei ole tapahtunut. 81 Kaikki KUUKAUSI JouMarLokSyyEloHeiKesTouHuhMaaHelTam V al ue 800 700 600 500 400 300 200 100 0 Kaikki 85 Kaikki 85 Kotimaa 85 Kotimaa 97 Ulkomaa 85 Ulkomaa 97 Kuvio 2. Matkustuskuukausi-indeksit EU-12 maissa Kausiluonteisuuden muutokset ovat varsin vähäisiä ja valtaosan niistä voitaneen tulkita johtuvan jokavuotisesta vaihtelusta. Toisaalta esimerkiksi ulkomaille tehtyjen vuoden päämatkojen kohdalla kausiluonteisuus on kasvanut varsin paljon Ranskassa ja Portugalissa, joskin jälkimmäistä selittänee matkustaneiden vähäisyydestä johtuvat satunnaiset vaihtelut. Laskua kausiluonteisuudessa on ollut erityisesti Tanskassa ja Alankomaissa, joissa se näkyy kaikkien matkojen ja ulkomaanmatkojen kohdalla. Muutoksen merkittävyyttä on kuitenkin varsin vaikea arvioida johtuen tarkasteluajankohtien vähäisyydestä. 4.6 Kausiluonteisuuden muutos Aineiston tuloksia tulkittaessa on syytä huomioida, että uusinkin tutkimus on tehty kuusi vuotta sitten. Kausiluonteisuus on kuitenkin varsin vakiintunut ilmiö ja sen vuosittaiset muutokset ovat varsin vähäiset. Aineiston avulla oletettavasti voidaan tarkastella mahdollisia suurehkoja kausiluonteisuuden muutoksia ja toisaalta vertailla eri maiden välisiä eroja. Mitään suurehkoja muutoksia ei kuitenkaan löydetty ja valtaosa matkoista on tehty perinteisinä loma-aikoina. Uusia varsinaisia sesonkeja ei ole ilmaantunut, joskin osassa maita pääsesongin kuukausi oli vaihtunut. Isossa- Britanniassa matkailun kausiluonteisuus oli kaikkein vähäisintä. Kausiluonteisuusindikaattoria ja gini-kerrointa tarkastelemalla 82 kausiluonteisuuden muutosta ei kuitenkaan ollut juurikaan havaittavissa. Sen sijaan maiden väliset erot olivat varsin isot. Tuloksesta voidaan epäsuorasti päätellä, että matkojen ajankohta on yhä varsin riippuvainen vuosilomien ajankohdista. Matkailuelinkeinon kannalta tämä tarkoittaisi, että kausiluonteisuus säilynee ongelmana jatkossakin, joskin erimaalaisten matkailijoiden riippuvuus traditionaalisista loma-ajoista on jonkin verran erilainen. Tämän perusteella kausiluonteisuuden ei voi olettaa katoavan, mutta hyödyntämällä erimaalaisten matkailijoiden erilaisia lomailuaikoja on suosituimmissa kohteissa mahdollista pidentää sesonkia. Kausiluonteisuutta mitattiin kahdella erilaisella mittarilla, kausiluonteisuusindikaattorilla ja gini-kertoimella, joiden havaittiin vastaavan toisiaan kohtuullisesti. Kausiluonteisuusindikaattorin ongelmana on kuitenkin sen voimakas riippuvuus suosituimman kuukauden arvosta. Se on oletettavasti varsin käypä mittari ymmärrettävyytensä puolesta, kun kyse on esimerkiksi hotelliyöpymisistä, mutta mikäli matkailijamäärillä ei ole kuukausittaista ylärajaa, gini-kerrointa voidaan pitää suositeltavana. Lisäksi havaittiin kausiluonteisuuden graafisen kuvaamisen merkitys. Esimerkiksi suosituimman kuukauden vaihtumista ei olisi havaittu kausiluonteisuusindikaattorista tai gini-kertoimesta. 83 Kirjallisuus Bardolet, E. (1999) A Introduction: Economic, Political and Social Aspects. Julkaisematon esitelmä. European North-South Tourism Education Bridge. Palma de Mallorca. Baum, T. (1999). Introduction. Tourism Economics, Vol. 5, No. 1, 5–8. Bell, D. (1974) The Coming of the Post-Industrial Society: A Venture in Social Forecasting. London: Heinemann. Butler, R. (1994) Alternative Tourism: The thin edge of the wedge. Teoksessa: Tourism Alternatives. Potentials and problems in the development of tourism, toim. V. Smith ja W. Eadington, 31–46. John Wiley & Son: Chichester. Edgar, D. (2001) Short Breaks Markets – from product positioning to a value- based approach. Teoksessa: Tourism in the 21st Century. Lessons from experience, toim. B. Faulkner, G. Moscardo ja E. Laws, 67– 88. Continuum: London & New York. Featherstone, M. (1995) Undoing Culture. Globalization, Postmodernism and Identity. Sage: London, Thousand Oaks & New Delhi. Frechtling, D. (1996) Practical Tourism Forecasting. Butterworth Heinemann: Oxford. Giles, D. (2002) Calculating a Standard Error for the Gini Coefficient: Some Further Results. Working Paper EWP0202, Department of Economics, University of Victoria: Victoria Hartman, R. (1986) Tourism, Seasonality and Social Change. Leisure Studies, Vol. 5, No. 1, 25–33. Honkanen, A. (2000) Seasonal Differences and Similarities in Tourism between Finland and Spain in the 1990s. Discussion and Working Papers Series No. 2, The Finnish University for Tourism Studies: Savonlinna. Kennedy, E. – Deegan, J. (2001) Seasonality in Irish Tourism 1973-1995. Teoksessa: Seasonality in Tourism, toim. T. Baum ja S. Lundtorp, 51–74. Pergamon: Oxford. Koenig, N. – Bischoff, B. (2003) Seasonality of Tourism in Wales: A Comparative Analysis. Tourism Economics, Vol. 9, No. 3, 229– 254. Lundtorp, S. (2001) Measuring Tourism Seasonality. Teoksessa: Seasonality in Tourism, toim. T. Baum ja S. Lundtorp, 23–50. Pergamon: Oxford. 84 Lundtorp, S. – Rassing, C. – Wanhill, S. (1999) The Off-Season Is ‘No Season’. The case of the Danish Island of Bornholm. Tourism Economics, Vol. 5, No. 1, 49–62. McEniff, J. (1992). Seasonality of tourism Demand in the European Community. Travel and Tourism Analysis, Vol. 3, No. 1, 67–88. Melich, A. (1999) Eurobarometer 48,0: Holiday travel, October-November 1997. [Elektroninen aineisto].Brussels: INRA (Europe) [tuottaja], Ann Arbor, MI: Inter-university Consortium for Political and Social Research; Köln: Zentralarchiv für Empirische Sozialforschung; Yhteiskuntatieteellinen tietoarkisto [jakajat]: Tampere. Melich, A. (2002) Eurobarometer 48,0: Holiday travel, October-November 1997. Codebook. Inter-university Consortium for Political and Social Research. (Saatavilla internetissä http://www.za.unikoeln.de/data/en/eurobarometer/codebooks/s299 cdb.pdf. (Viitattu 15.11.2002)). Miles, S. (1998) Consumerism – as a Way of Life. Sage: London, Thousand Oaks & New Delhi. Moran, T. (2003) On the Theoretical and Methodological Context of Cross- National Inequality Data. International Sociology, Vol. 18, No. 2, 351–378. Rabier, J-R. – Riffault, H. – Inglehart, R. (1986). Eurobarometer 25 : holiday travel and environmental problems, April 1986 [Elektroninen aineisto]. Faits et Opinions, Paris [aineistonkeruu]. ICPSR ed. Ann Arbor, MI : Inter-university Consortium for Political and Socia Research [tuottaja ja jakaja], 1988. Yhteiskuntatieteellinen tietoarkisto [jakaja]: Tampere. Rojek, C. 2000. Leisure and Culture. MacMillan: Hamshire & London. Standard Eurobarometer Weighting Overview (2003). (Saatavilla internetissä http://www.gesis.org/en/data_service/eurobarometer/standrd_eb/e bweight.html. (Viitattu 15.11.2003)). Syrjämaa, T. (2001). Challenging Seasons: Pros and Cons of Changing Seasonality Patterns in Early 20th Century Europe. Teoksessa: North – South: Contrasts and Connections in Global Tourism, toim. T. Toivonen ja A. Honkanen, 69–79. Proceedings of 7th ATLAS international Conference, ATLAS & Finnish University Networks for Tourism Studies: Savonlinna. Toivonen, T. (1999). Seasonality – Only a Disadvantage for Tourism Industry? Teoksessa: Tourism Industry and Education Symposium, Jyväskylä, Finland, September 23-26, 1999, Toim. E. 85 Arola & T. Mikkonen, 128–135, Jyväskylä Polytechnic: Jyväskylä. Wanhill, S. (1980). Tackling Seasonality. A technical note. International Journal of Tourism Management Vol 1, No. 4, 243–245. Yacoumis, J. (1980). Tackling Seasonality - The case of Sri Lanka. International Journal of Tourism Management Vol. 1, No. 2, 84– 98. 86 Liite 1. Matkustuskuukausi-indeksit maittain Iso-Britannia KUUKAUSI JouMarLokSyyEloHeiKesTouHuhMaaHelTam V a l u e 800 700 600 500 400 300 200 100 0 Kaikki 85 Kaikki 85 Kotimaa 85 Kotimaa 97 Ulkomaa 85 Ulkomaa 97 Tanska KUUKAUSI JouMarLokSyyEloHeiKesTouHuhMaaHelTam V a l u e 800 600 400 200 0 Kaikki 85 Kaikki 85 Kotimaa 85 Kotimaa 97 Ulkomaa 85 Ulkomaa 97 Saksa KUUKAUSI JouMarLokSyyEloHeiKesTouHuhMaaHelTam V a l u e 800 700 600 500 400 300 200 100 0 Kaikki 85 Kaikki 85 Kotimaa 85 Kotimaa 97 Ulkomaa 85 Ulkomaa 97 Kreikka KUUKAUSI JouMarLokSyyEloHeiKesTouHuhMaaHelTam V a l u e 800 700 600 500 400 300 200 100 0 Kaikki 85 Kaikki 85 Kotimaa 85 Kotimaa 97 Ulkomaa 85 Ulkomaa 97 Espanja KUUKAUSI JouMarLokSyyEloHeiKesTouHuhMaaHelTam V a l u e 800 700 600 500 400 300 200 100 0 Kaikki 85 Kaikki 85 Kotimaa 85 Kotimaa 97 Ulkomaa 85 Ulkomaa 97 Ranska KUUKAUSI JouMarLokSyyEloHeiKesTouHuhMaaHelTam V a l u e 800 700 600 500 400 300 200 100 0 Kaikki 85 Kaikki 85 Kotimaa 85 Kotimaa 97 Ulkomaa 85 Ulkomaa 97 87 Irlanti KUUKAUSI JouMarLokSyyEloHeiKesTouHuhMaaHelTam V a l u e 800 700 600 500 400 300 200 100 0 Kaikki 85 Kaikki 85 Kotimaa 85 Kotimaa 97 Ulkomaa 85 Ulkomaa 97 Italia KUUKAUSI JouMarLokSyyEloHeiKesTouHuhMaaHelTam V a l u e 800 600 400 200 0 Kaikki 85 Kaikki 85 Kotimaa 85 Kotimaa 97 Ulkomaa 85 Ulkomaa 97 Luxemburg KUUKAUSI JouMarLokSyyEloHeiKesTouHuhMaaHelTam V a l u e 800 700 600 500 400 300 200 100 0 Kaikki 85 Kaikki 85 Ulkomaa 85 Ulkomaa 97 Alankomaat KUUKAUSI JouMarLokSyyEloHeiKesTouHuhMaaHelTam V a l u e 800 700 600 500 400 300 200 100 0 Kaikki 85 Kaikki 85 Kotimaa 85 Kotimaa 97 Ulkomaa 85 Ulkomaa 97 Portugali KUUKAUSI JouMarLokSyyEloHeiKesTouHuhMaaHelTam V a l u e 800 600 400 200 0 Kaikki 85 Kaikki 85 Kotimaa 85 Kotimaa 97 Ulkomaa 85 Ulkomaa 97 Belgia KUUKAUSI JouMarLokSyyEloHeiKesTouHuhMaaHelTam V a l u e 800 700 600 500 400 300 200 100 0 Kaikki 85 Kaikki 85 Kotimaa 85 Kotimaa 97 Ulkomaa 85 Ulkomaa 97 88 89 5 AIKA, SOSIAALINEN KAUSAATIO JA RAKENTEELLISET VAIKUTUKSET VERTAILEVASSA TUTKIMUKSESSA Juho Härkönen Tiivistelmä Aikaan liittyviin analyyttisiin kysymyksiin on sosiaalitieteissä kiinnitetty varsin vähän huomiota. Tässä artikkelissa tarkastellaan näitä kysymyksiä, keskittyen vertailevaan tutkimukseen, jossa ajan käsittelemiseen liittyvää keskustelua on käyty etenkin historiallisen vertailevan tutkimuksen ympärillä. Artikkelissa asetutaan kannalle, jonka mukaan aikaan liittyvien analyyttisten erityiskysymysten tarkempi huomioon ottaminen sekä teorioissa että metodologiassa edesauttaisi sosiologista tutkimusta yleensä, ja kausaatiopäätelmiin pyrkivää tutkimusta erityisesti. ASIASANAT: aika, kausaatio, metodologia, teoria 5.1 Johdanto Sosiologisessa kirjallisuudessa esiintyvä aikaan liittyvien kysymysten eksplisiittinen pohdinta on perinteisesti liittynyt sosiologian ja historiatieteen välisen suhteeseen. Näissä keskusteluissa keskeisenä teemana on ollut sosiologian historiattomuuden kritisointi: sosiologit eivät tunne historiaa tarpeeksi, eivätkä kiinnitä tarpeeksi huomiota historiallisiin prosesseihin, ja niiden merkitykseen nyky-yhteiskuntia muokkaavina tekijöinä. Eräs tunnetuimmista sosiologian historiattoman luonteen kritisoijista on C. Wright Mills, joka arvosteli kärjekkäästi sen aikaisia teoreetikkoja sekä etenkin kvantitatiivis-empiirisesti suuntautuneita tutkijoita klassiseen sosiologiaan erottamattomasti kuuluneen historiallisen lähestymistavan hylkäämisestä (Mills 1970). Vastaavanlaista kritiikkiä esitetään nykyään varsinkin niin kutsuttujen historiallisten sosiologien piirissä (esim. Skocpol 1984; Ruonavaara 1995). Historiattomuuden lisäksi sosiologiaa on yleisemmin kritisoitu ajan ja siihen liittyvien teoreettisten ja metodologisten kysymysten unohtamisesta 90 (Abbott 2001). Sosiologiset teoriat ja menetelmät eivät tämän kritiikin mukaan kiinnitä tarpeeksi huomiota sosiaalisten ilmiöiden ajalliseen luonteeseen, aikajärjestykseen ja ajoitukseen. Tämä heikentää mahdollisuuksiamme kausaalisten mekanismien analysointiin, sosiaalisten prosessien ymmärtämiseen ja teorioiden testaamiseen (esim. Blossfeld 1996; Blossfeld – Rohwer 2001). Sosiologisesti mielenkiintoiset ilmiöt ja yhteydet ovat aina kehittyneet ajan myötä. Staattisilla teorioilla ja menetelmillä näiden prosessien tutkiminen ei ole mahdollista, mikä puolestaan vaikuttaa tutkittavien ilmiöiden ymmärtämiseen. Nancy Tuma ja Michael Hannan (1984, 12), jotka ovat pitkään peräänkuuluttaneet sosiaalisen dynamiikan huomioimisen tärkeyttä, ovat huomauttaneet, että staattisilla menetelmillä saamme usein käsityksen, että kaikki vaikuttaa kaikkeen. Dynaamisella analyysilla voidaan sen sijaan tarkastella ilmiöiden ja havaittujen yhteyksien kehittymistä ajassa, ja tarkastella erilaisia ilmiöiden kehittymiseen vaikuttavia tekijöitä. Hans-Peter Blossfeld puolestaan väittää jopa, että ajan tarkempi huomioonottaminen voi auttaa sellaisten perisosiologisen kysymyksen kuin rakenteen ja toiminnan välisen suhteen selventämisessä. Keskeistä on huomata, että jos rakenteet vaikuttavat toimintaan, tai toisin päin, näillä vaikutuksilla on aina ajallinen luonteensa. Tässä artikkelissa pohditaan ajan analyyttista merkitystä, keskittyen ajan merkitykseen sosiologisia kausaalipäätelmiä tehtäessä. Alun yleisemmän tason keskustelun jälkeen tarkastelu keskittyy vertailevan tutkimuksen kontekstiin. Vertaileva tutkimus on sikäli hedelmällinen tarkastelun kohde, että siinä aikaan liittyvien kysymysten problematiikka on noussut esiin historiallisen vertailevan tutkimuksen, ja sen metodologisten kysymysten kautta. Tarkastelu rajautuu myöhemmin vielä tarkemmin tutkimukseen, jossa analysoidaan makrotason rakenteiden vaikutusta mikrotason toimintaan. Aikaan ja ajallisuuteen liittyvien kysymysten huomioon ottaminen on erityisen kiinnostava ja haastava aihe tämän sosiologisesti klassisen kysymyksen pohdinnassa. Artikkelin kysymykset nousevat pitkälti vertailevasta tutkimuksesta. Aihetta käsitellään lähinnä kvantitatiivisen tutkimuksen näkökulmasta, joskin samat teemat ovat relevantteja myös muunlaisen tutkimuksen kannalta. 5.2 Aika ja kausaaliset mekanismit Sosiologisten kausaalipäätelmien tekemistä on viime vuosikymmeninä kritisoitu ja epäilty laajalti. Osittain tämä liittyy yleisempiin teoreettisiin ja filosofisiin pohdintoihin sosiaalisen ja sosiologisen tutkimuksen luonteesta, toisaalta taas metodologisiin huomioihin tyypillisen sosiologisen tutkimuksen 91 ja kokeellisen tutkimuksen eroista (taloustieteen osalta ks. LaLonde 1986). Tästä huolimatta syy-seuraussuhteiden esittäminen eri sosiaalisten tekijöiden välillä on yleistä. Sosiaalista kausaatiota koskevissa teoreettisissa keskusteluissa on viime aikoina painotettu näkemystä, jonka mukaan kausaatiovaikutusten analysoimiseksi tutkimuksissa tulisi tehdä eksplisiittisiä hypoteeseja kausaatiota tuottavista sosiaalisista mekanismeista (Hedström – Swedberg 1998). Keskustelu sosiaalisista mekanismeista paitsi tuo pohdinnan sosiaalisesta kausaatiosta vähemmän abstraktille ja konkreettisesti testattavammalle ”middle-range” tasolle, niin myös luo tärkeän sillan lähinnä kokeellisen tutkimuksen parissa kehitettyjen kausaalisuuden teorioiden ja pitkälti ei-kokeelliseen tutkimukseen perustuvan sosiologian välillä. Kausaatiomallien kirjallisuudessa (esim. Holland 1986) lähdetään yleisesti liikkeelle oletuksesta, että kausaatiota voidaan tulkita ainoastaan kahden tapahtuman, siis muuttujassa tapahtuvan muutoksen, pohjalta. Ajan myötä pysyvien muuttujien tapauksessa kausaatiota ei voida tulkita. Esimerkiksi työmarkkina-aseman muutos voi kausaalisesti vaikuttaa köyhyyteen, kun taas sukupuoli ei voi olla köyhyyden syy. Blossfeld ja Rohwer (2001) kuitenkin huomauttavat, että vaikka (tutkimusaikavälin) hetkellä muuttumattomia tekijöitä ei tiukan kausaalisessa mielessä voida pitää minkään tapahtuman syynä, voidaan niitä kuitenkin (varovasti) käsitellä kausaalisina tekijöinä. Ideana tässä on, että tämän tyyppiset tekijät ovat sosiaalisesti merkityksellisinä toimintaa suuntaavia. Tämä oletus vaatii taustalleen teoreettisen näkemyksen kyseisten tekijöiden vaikutuksista. Esimerkiksi sukupuolten välisiä eroja työmarkkinoilta pois jäämisessä lapsen syntymän jälkeen voidaan tulkita sukupuolesta johtuviksi: sukupuoli on ikään kuin havainnoimattomien tekijöiden (kuten sosialisaatiossa opittujen roolien) aggregaatti ja likiarvo. Ilman teoriaa näitä tekijöitä yhdistävistä sosiaalisista mekanismeista tällaisen tulkinnan tekeminen olisi mahdotonta1. Sosiaalisista mekanismeista ja niiden avulla suoritettavasta kausaalisesta päättelystä käyty keskustelu on tuonut esiin useita tärkeitä tekijöitä. Kuitenkin myös siihen pätee pitkälti Abbottin (2001) kritiikki varsin vähäisestä aikaan liittyvien kysymysten huomioimisesta sosiologisissa teorioissa. Kausaatiota käsittelevässä metodologisessa kirjallisuudessa aikaan sen sijaan on kiinnitetty enemmän huomiota. Eniten huomiota on saanut tapahtumien välisen ajallisen järjestyksen tarkastelu. Jotta kahden tapahtuman välille voidaan tulkita 1 Teoria kausaatiovaikutuksen tuottavista sosiaalisista mekanismeista mahdollistaa siis kausaatiotulkintojen tekemisen ei-kontrolloiduissa tilanteissa. Näissäkin tapauksissa tulee luonnollisesti huomata mahdollisten havainnoimattomien muuttujien väliin tuleva vaikutus. Kontrolloimattomissa ja satunnaistamattomissa asetelmissa kausaatiopäätelmien teko on loppujen lopuksi aina ongelmallista. 92 kausaatiosuhde, täytyy syyn edeltää seurausta. Tämä on jokaisessa kausaatiota käsittelevässä tekstissä usein esiintuotu seikka. Toinen, joskin vähemmän huomiota saanut kausaalisten mekanismien vaikutusten ja ajan välistä suhdetta koskeva tekijä on kunkin kausaalisen mekanismin aikaansaaneen tapahtuman ajoitus ja tapahtumien välinen viive, ja kausaalisen mekanismin funktionaalinen muoto ajan suhteen (Blossfeld – Rohwer 2001). Viiveellä tarkoitan tietyn tapahtuman ja sen vaikutuksen välistä aikaa. Millään tapahtumalla ei ole samanhetkistä vaikutusta, vaan vaikutuksen ilmenemiseen kuluu aina tietty määrä aikaa. Työttömyys ei johda köyhyyteen sillä hetkellä kun henkilö kävelee työpaikastaan pois viimeisen kerran, vaan köyhyys (jos on tullakseen) seuraa työttömyyttä tietyllä viiveellä. Jälkimmäisellä puolestaan viitataan tapahtuman “intensiteetin” kehittymiseen ajan myötä. Edellistä esimerkkiä seurataksemme voimme ajatella, että työttömyys lisää köyhyysriskiä ajan myötä tasaisesti, hyppäyksenomaisesti tietyn ajan päästä, tai kiihtyvällä vauhdilla (ks. kuvio 1). Tapahtumien välisen viiveen ja kausaalisen mekanismin funktionaalisen suhteen huomioon ottaminen täsmentää tietämystämme tarkasteltavista kausaalisuhteista. Puhtaasti poikkileikkausasetelmaan nojautuva tutkimus saattaa sen sijaan antaa siitä harhaista informaatiota. c b a t Kuvio 1. Ajan suhteen a) tasainen, b) hyppäyksenomainen ja c) kiihtyvästi kehittyvät kausaatiovaikutukset. 93 Tästä voimme ottaa äärimmäisen esimerkin. Oletetaan, että tutkittaessa työttömyyden vaikutusta köyhyyteen poikkileikkausaineistolla saamme tuloksen, jonka mukaan työttömyys ja köyhyys eivät ole yhteydessä keskenään. Sen sijaan, että tämä yhteys olisi oikeasti olemassa, tämän havainnon on saattanut tuottaa kaksi eri tekijää. Ensinnä voimme (kärjistetysti) olettaa tilanteen, jossa aineisto on kerätty tilanteessa, jossa kukaan työtön ei ole vielä pudonnut köyhyysrajan alle2. Toiseksi tulos voi syntyä kahden vastakkaisen efektin toimesta. Poikkileikkausaineistolla saatava estimaatti kuvaa työttömyyden nettovaikutusta tietyllä hetkellä (Tuma – Hannan 1984; Blossfeld – Rohwer 2001). Tällöin on mahdollista, että työttömyys kaksinkertaistaa köyhäksi joutumisen riskin kullakin hetkellä (nopeuttaa köyhtymistä), mutta myös kaksinkertaistaa mahdollisuuden köyhyydestä poistumiseen kullakin hetkellä (nopeuttaa köyhyydestä poistumista). Tällöin poikkileikkausasetelmalla havaittava nettovaikutus olisi kuitenkin nolla. Työmarkkina-asemalla on siis todellisuudessa vaikutus, jota ei kuitenkaan poikkileikkausasetelmalla havaittaisi. Jälkimmäisessä tapauksessa onkin tärkeä huomata, monia sosiologisesti kiinnostavia tekijöitä voidaan hedelmällisesti tarkastella muutoksen ja pysyvyyden vuorovaikutuksena. Näihin kysymyksiin liittyvän problematiikan huomioiminen, on lisännyt kvantitatiivisin menetelmin tehtävää sosiologisten kysymysten ajallista tutkimusta. Tämän on mahdollistanut myös sekä prospektiivisten että retrospektiivisten aineistojen lisääntynyt saatavuus, analyysimenetelmien ja – ohjelmistojen kehittyminen ja tietokoneiden laskentatehon parantuminen. Usein pitkittäisasetelmalla tehdyt tutkimukset ovatkin kumonneet poikkileikkausaineistoin saatuja näkemyksiä. Oppenheimer (1997) kävi läpi tutkimuksia, joissa testattiin hypoteesia naisten lisääntyneen koulutuksen ja työssäkäynnin, ja vähentyneen avioitumisen välillä. Hänen johtopäätöksensä oli, että koulutuksen ja työssäkäynnin, ja avioitumisen välinen negatiivinen yhteys sai tukea ainoastaan poikkileikkauksin tehdyistä tutkimuksista. Tämä johtui siitä, että näissä tutkimuksissa avioituminen ja avioitumisen viivästyminen sekoittuivat keskenään. Naisten koulutus ja työssäkäynti viivästytti avioliiton solmimista, mutta ei vähentänyt avioitumista (ks. myös Blossfeld – Huinink 1991). Viivästynyt avioituminen puolestaan näyttäytyi poikkileikkausaineistoissa naisten koulutuksen ja työssäkäynnin, ja avioitumisen negatiivisena yhteytenä. 2 Tätä lähellä oleva tilanne voisi syntyä työttömyysasteen rajusti kohotessa. 94 5.3 Metodologiset ratkaisut ja aika vertailevassa makrososiologiassa Esimerkkinä ajan roolin käsittelemisestä osana tutkimuksen metodologista asetelmaa käytän vertailevan tutkimuksen parissa käytyä, välillä kärjekästäkin keskustelua vertailevan tutkimuksen metodologisista ratkaisuista. Eräs tämän keskustelun teema on koskenut historiallisten prosessien huomiointia nykyhetken selittäjinä. Keskeisimmin tämä keskustelu on koskenut vertailevaa makrososiologista tutkimusta, jossa mielenkiinnon kohteena ovat esimerkiksi poliittiset järjestelmät (Berg-Schlosser – de Meur 1994), vallankumoukset (Skocpol 1979) tai hyvinvointivaltiot (Esping-Andersen 1990). Toisella puolella ovat olleet kvalitatiivisesti orientoituneet tutkijat, jotka ovat painottaneet tutkittavien kohteiden holistisen ja erityislaatuisen luonteen korostamista kohteiden ja niiden kehityskulkujen oikeanlaisen ymmärtämisen kannalta (ks. esim. Ragin 1987; 2000). Toisella puolella ovat puolestaan olleet kvantitatiivisesti orientoituneet tutkijat, jotka ovat lähteneet liikkeelle yleisten teorioiden testaamisesta tilastollisin menetelmin (esim. Przeworski – Teune 1970). Tämän paperin puitteissa ei ole mahdollista lähteä kartoittamaan kaikkea kyseiseen vastakkainasetteluun liittyvää keskustelua (tätä varten ks. Ragin 1987; 2000; Goldthorpe 1997, ja tätä kommentoivat artikkelit teoksessa Mjøset 1997), vaan aion keskittyä vertailevan tutkimuksen teorian ja metodologian tarkasteluun aikaan liittyvien analyyttisten kysymysten näkökulmasta. Eräs kirjallisuudessa esiin nostettu, ja konkreettisesti aikaan ja ajallisuuden käsittelemiseen liittyvä vertailevan tutkimuksen ongelma liittyy niin kutsuttuun “mustan laatikon ongelmaan” (Goldthorpe 1997; Abbott 2001). Etenkin kvantitatiivisesti orientoitunutta tutkimusta on kritisoitu liiasta abstraktisuudesta ja muuttujien taustalla operoivien aktuaalisten sosiaalisten prosessien huomioimattomuudesta. Tätä “muuttujasosiologian” (Esser 1996) ongelmaa on kritisoitu siitä, että se pahimmillaan olettaa muuttujien olevan toimijoiden sijaan sosiaalisen muutoksen agentteja. Kritiikki on siis ollut samankaltaista kuin sosiaalisten mekanismien puolestapuhujien esittämä. Vertailevan tutkimuksen(kin) kohdalla tähän ongelmaan on tarjottu ratkaisuksi kvalitatiivisemmin orientoitunutta tutkimusotetta, jossa itse lopputuloksen prosessia tarkastellaan yksityiskohtaisen informaation ja holistisen näkökulman avulla. Tämän kannan mukaan holistista näkökulmaa tarvitaan, koska sosiaalisia prosesseja ei voida järkevästi tulkita toisistaan riippumattomiksi oletettujen tekijöiden summana. Eräänä, prosessien ajallisen komponentin huomioon ottavana vaihtoehtona on tarjottu vertailevaa historiallista tutkimusta (esim. Rueschemayer – Stephens 1997). Tällaisella lähestymistavalla, jossa toimijuus ja historialliset tapahtumaketjut otetaan eksplisiittisen tarkastelun kohteeksi, voidaan sen kannattajien mukaan paitsi 95 kuvata näitä kehityskulkuja, niin myös päästä parempiin johtopäätöksiin kilpailevien selitysten osalta (emt., 61). Vaikka mustan laatikon ongelmaa, ja siihen liittyvää aktuaalisten ja teoreettisten prosessien vähäistä huomioimista on pidetty lähinnä kvantitatiivisen tutkimuksen ongelmana, esimerkiksi Goldthorpe (1997) on väittänyt, että nämä ongelmat eivät ole mitenkään leimallisesti kvantitatiiviseen orientaatioon liittyviä. Rueschemayer ja Stephens kritisoivatkin prosessien huomioimattomuudesta paitsi tilastollisia menetelmiä, niin myös QCA:ta käyttäviä tutkimuksia (Rueschemayer – Stephens 1997). Onkin huomionarvoista, että esimerkiksi Charles Raginin lanseeraamissa QCA (1987), että myös “edistyneemmässä”, sumeita joukkoja käyttävässä FsQCA (2000) -menetelmissä painotetaan prosessien ajallista aspektia varsin vähän. Vaikka prosessit ovat mukana Raginin teoreettisessa kielenkäytössä, itse analyysimenetelmät ovat pitkälti luonteeltaan poikkileikkausten analyysiin tarkoitettuja. Lopputuloksia (esim. vallankumous) selitetään eri selittävien muuttujien kombinaatioilla, mutta selitettävien muuttujien / tapahtumien ajalliselle järjestykselle ei juurikaan panna painoa, kuin ei myöskään selittävien kombinaatioiden kausaalisen mekanismin ja sen tuloksen aikavälille ja kausaalisten mekanismien “funktionaaliselle muodolle”3 ajan suhteen. Stefano Bartolini on esittänyt vastaavaa kritiikkiä myös avoimesti historiallista sosiologiaa kohtaan. Bartolinin mukaan kyseinen koulukunta vetoaa usein intuitiivisesti helposti yhdyttävissä oleviin näkemyksiin historiallisten prosessien tärkeydestä, mutta ei kiinnitä metodologisissa ratkaisuissaan riittävää huomiota aikaan liittyviin erityiskysymyksiin, vaan käsittelee ajankohtien vertailua samoin kuin paikkojen vertailua. Hän onkin peräänkuuluttanut pikemmin aikaan kuin historiaan keskittyviä tutkimusmenetelmiä. (Bartolini 1993.)4 Vertailevaan kvantitatiiviseen tutkimukseen liittyen ja edellä käytyyn keskusteluun liittyen onkin helppo ajatella, että sosiaalisen dynamiikan parempi huomioon ottaminen kvantitatiivisessa vertailevassa makrososiologisessa tutkimuksessa voi auttaa mustalaatikoinnin vähentämisessä5. Tällöin päästäisiin lähemmäs muun muassa Rueschemayerin ja Stephensin vaalimaa historiallista vertailevaa tutkimusta samalla pitäen kiinni tilastollisen mallinnuksen eduista. Tämä voisi auttaa myös teorian ja aineiston yhteensovittamisessa, etenkin jos teoreettisiksi lähtökohdiksi otetaan 3 Käytän tässä lainausmerkkejä, koska kausaalisten mekanismien ja niiden vaikutusten ajan funktionaalisuuteen liittyvä kielenkäyttö liittyy keskeisesti tilastollisten menetelmien käyttöön, kun taas Ragin tunnetusti pitää menetelmiään vaihtoehtoina tilastolliselle mallinnukselle. 4 Kuitenkin myös kvalitatiivisen tutkimuksen piirissä on kehitetty ajan eksplisiittisesti huomioivia menetelmiä, esimerkkinä Event Structure Analysis (Toivonen 1999, 159–167). 5 Tämä ei kuitenkaan (välttämättä) lievennä kvantitatiivisen vertailevan tutkimuksen perisyntiä, eli tapausten vähäistä lukumäärää. 96 eksplisiittisiä oletuksia makrotason kehityskulkujen aikajärjestyksestä ja tapahtumien ajoitukseen liittyvistä tekijöistä. Tämäntyyppistä kvantitatiivisesti orientoitunutta vertailevaa tutkimusta onkin tehty (ks. Esping-Andersen – Przeworski 2002). Esimerkiksi Kangas (2003) käytti eksplisiittisesti aikaan sidottuja menetelmiä tutkiessaan eri tekijöiden vaikutuksia sosiaaliturvalainsäädännön ajoitukseen. Ajoituksen huomioon ottavalla tutkimusasetelmalla voitiin saada tukea hypoteeseille, joiden testaaminen poikkileikkausasetelmalla olisi ollut hankalaa. 5.4 Aika, ja siirtyminen makrosta mikroon Edellä esitelty keskustelu on koskenut pitkälti makrososiologista vertailevaa tutkimusta. Keskustelua voidaan kuitenkin laajentaa myös mikrotason ilmiöistä kiinnostuneeseen vertailevaan tutkimukseen, jossa tarkastellaan makrotason vaikutuksia alemman tason (yksilöiden, ryhmien jne.) toimintaan. Tällaiselle tutkimukselle on tyypillistä, että maita ei tutkita ainoastaan niiden itsensä vuoksi, vaan tutkimuksen avulla pyritään usein testaamaan yleisemmän tason hypoteeseja makrotasoa alemman tason toiminnasta (Kohn 1989, 21– 22). Tällaiselle vertailevalle tutkimukselle on siis tyypillistä, että kohdetta analysoidaan useammalla kuin yhdellä tasolla (Przeworski – Teune 1970, 50- 51)6. Useammalla tasolla tapahtuvaa sosiologista analyysia voidaan myös käyttää makrotason ilmiöiden tutkimiseen, ja eräiden mielestä tällaista lähestymistapaa tulisikin käyttää makrososiologisessa tutkimuksessa (Coleman 1990). Tätä lähestymistapaa, jossa sosiaalisia ilmiöitä pyritään pohjimmiltaan selittämään yksilötason toiminnasta, kutsutaan metodologiseksi individualismiksi7. Useamman tason sosiologisessa analyysissa keskeistä on se, miten eri tasot yhdistetään teoreettisesti (Coleman 1990). Colemanin mukaan kyseessä on makro-mikro-makro –siirtymien toteuttaminen analyyttisesti kestävällä tavalla (nk. siltahypoteesit (esim. Esser 1998)). Colemanin mukaan tällaisen sosiologisen selityksen tulisi sisältää a) analyyttisen siirtymisen makrosta mikroon (makrotason tekijöiden vaikutus yksilöiden toimintaympäristöön ja orientaatioihin), b) teorian yksilöiden toiminnasta, ja c) analyyttisen siirtymisen mikrosta makroon (yksilöiden toiminnan aggregaattitason 6 Useammalla tasolla operoivaa analyysia varten on myös kehitetty omia tilastollisia analyysimenetelmiään (multilevel analysis) (esim. DiPrete – Forristal 1994; Snijders – Bosker 1999). 7 Metodologinen individualismi on suosittua etenkin rationaalisen valinnan (toiminnan) teoreetikoiden parissa. Tulee kuitenkin huomata, että metodologinen individualismi ei ole sidottu rationaalisen valinnan (toiminnan) teoriaan, vaan sitä voidaan soveltaa laajemmin. Onkin väitetty, että kaikki parhaat sosiologiset selitykset pohjautuvat metodologiseen individualismiin, vaikkakin ehkä ainoastaan implisiittisesti (Boudon 2002). 97 seuraukset)8. Vertaileva sosiologinen tutkimus, joka hyväksyy metodologisen individualismin lähtökohdakseen, joutuu ottamaan kantaa vähintään toiseen analyyttiseen siirtymään (joko makro-mikro tai mikro-makro), ja usein kumpaankin. Tässä keskityn tarkastelemaan makro-mikro –siirtymien toteuttamista, ja ajan huomioimista makrotason elementtejä sisältävissä mikrotason selityksissä. Makro-mikro –siirtymät ovat käytännössä analyyttisia oletuksia yksilöä yleisemmällä tasolla olevista (yhteiskunnallisista ja sosiaalisista) rakenteista ja konteksteista, ja siitä, miten nämä vaikuttavat yksilön toimintamalleihin. Rakenteilla voidaan ymmärtää niitä toiminnan puitteita (settings), jotka ovat, ainakin potentiaalisesti, yhteisiä kaikille kyseisessä sosiaalisessa järjestelmässä toimiville (Przeworski – Teune 1970, 53–56). Tällaisena voidaan pitää esimerkiksi kansallista koulutusjärjestelmää ja hyvinvointivaltiota. Kontekstilla puolestaan voidaan Przeworskia ja Teunea mukaillen ymmärtää tiettyjä yksilötasosta nousevia aggregaattitekijöitä, jotka suuntaavat yksilöiden toimintaa (emt. 56–57). Tällaisesta esimerkkinä voidaan pitää yksilöiden arvoympäristöä. Tiukan fundamentaalisessa ympäristössä vallitseva arvokonservativismi on tämän näkemyksen mukaan yksilöiden konservatiivisuuden ”summa”. Kyseinen arvoympäristö puolestaan luonnollisesti vaikuttaa yksilöiden toimintamalleihin. Przeworski ja Teune erottavat vielä kolmannen systeemitason analyyttisen tekijän, ”diffuusion” (diffusion patterns), joka viittaa esimerkiksi kulttuuriseen oppimiseen ja imitaatioon makrotasolla9. Tätä en tule kuitenkaan käsittelemään tämän pidemmälle. On selvää, että juuri makrotason vaikutusten tutkiminen on keskeistä vertailevan tutkimuksen kannalta: ideanahan on juuri erilaisten makrotasoisten systeemien vertailu. Edellä esitetyn pohdiskelun pohjalta on helppo ajatella, että aikaan liittyvien analyyttisten kysymysten huomioiminen on keskeistä myös vertailevan tutkimuksen kannalta. Jotta tietyllä yhteiskunnallisella rakenteella voidaan sanoa olevan vaikutusta yksilötason toimintaan, tulee kyseisen rakenteen edeltää toimintaa. Parhaiten rakenteen vaikutusta toimintaan voidaan tutkia, jos käytössämme on sellaisen aikavälin aineisto, jossa rakenteet ja kontekstit ovat muuttuneet tarkastelujakson aikana. Blossfeld (1996) korostaakin eri tasoilla tapahtuvien prosessien huomioimisen tärkeyttä. Hänen lähtökohtanaan pitämässään kausaalisuuden mallissahan (Blossfeld – Rohwer 2001) on keskeistä, että selittävä muuttuja (tässä 8 Analysoitavia tasoja voi luonnollisesti olla useampi kuin kaksi. Tällöin myös makro-mikro-makro – siirtymien lukumäärä lisääntyy. 9 Tässä on kyse nk. Galtonin ongelmasta, eli siitä, miten erottaa keskenään yhteiskunnan sisäisen kehityksen tuloksena syntyvät rakenteet, kulttuuriset muodot ja instituutiot muilta yhteiskunnilta matkituista. 98 makrotasolla oleva) muuttuu ennen selitettävää (mikrotason muuttuja). Vain näin voimme (kunnolla) päästä käsiksi kausaalisiin vaikutussuhteisiin. Mikroprosessi, muutos Makroprosessi, muutos t t1 Kuvio 2. Makrotason muutoksen vaikutus mikrotason prosessiin (tässä kiihtyvä kausaatiovaikutus). Usein käyttämämme aineistot eivät kuitenkaan ajallisesti riitä edellä esitetyn ideaalimallin tutkimiseen. Vaikka pitkittäisaineistoilla voimmekin usein tutkia yksilötason siirtymiä, aineisto ei aina ajallisesti riitä makrotason muutosten havaitsemiseen. Ajallisuuden huomioimisella on kuitenkin etunsa tässäkin tapauksessa. Ensinnäkin, kuten Abbott (2001) on muistuttanut, eräs rakenteiden keskeinen, joskin usein unohdettu piirre liittyy niiden pysyvyyteen alemman tason ilmiöihin verrattuna. Toiseksi, jos oletamme, että erilaisilla makrotason tekijöillä on vaikutusta mikrotason ilmiöihin, voimme myös olettaa, että tämä ero näyttäytyy mikrotason tapahtumien ajoituksessa ja niiden ajan suhteen toteutuvassa funktionaalisessa muodossa. Tässä mielessä ajan kulussa vakaita makrotason tekijöitä voidaan verrata aiemmin käsiteltyihin yksilötasolla (suhteellisen) pysyviin tekijöihin, kuten sukupuoleen. Makrotason tekijöiden kohdalla tarvitsemme yksilötason muuttumattomien tekijöiden tavoin teorian siitä, minkä mekanismien kautta makrotason tekijät vaikuttavat toimintaan, ja ajallista tietoa tapahtumista eri systeemeissä. Ajallisuuden merkitys puolestaan liittyy edellä käytyyn keskusteluun. Jos 99 vaikka systeemin A oletetaan vähentävän avioliittojen solmimista verrattuna systeemiin B, tämän tulisi näkyä siten, että systeemissä B avioliittoja solmitaan paitsi enemmän, niin myös nuoremmalla iällä kuin systeemissä A. Eräs käyttökelpoinen esimerkki tähän liittyen on McGinnityn (2002) tutkimus sosiaaliturvan vaikutuksista työttömien miesten vaimojen työllisyyteen. Aiempi tutkimus on osoittanut, että työttömien puolisoilla on tavanomaista suurempi todennäköisyys kokea työttömyyttä (esim. Henkens ym. 1993). On myös havaittu, että työttömyyden kasautumisen suhteen esiintyy maakohtaisia eroja (emt.; de Graaf – Ultee 2000). Eräänä syynä tähän on pidetty maiden sosiaaliturvassa esiintyviä eroja. Niissä järjestelmissä, joissa puolison tulot vaikuttavat saataviin tukiin, työttömien puolisot ovat eräänlaisessa kannustinloukussa: työllistyminen (tai työssä pysyminen) ei välttämättä lisää kotitalouden käteen jääviä tuloja, ja täten puolison on taloudellisesti kannattavampaa olla työttömänä. Poikkileikkausaineistolla tehdyt tutkimukset ovat tukeneet tätä hypoteesia. Esimerkiksi Dex ja kumppanit (1995) havaitsivat, että niissä maissa, joissa puolison tulot eivät pääsääntöisesti vaikuta saatuihin työttömyysavustuksiin, työttömyyden kasautuminen pariskuntiin on vähäisempää. Poikkileikkausaineistojen keskeisimpänä ongelmana voidaan kuitenkin pitää sitä, että niiden avulla ei ole mahdollista tutkia pariskunnittaisen työttömyyden esiintymiseen vaikuttavia prosesseja. On eri asia todeta, että työttömien puolisot ovat todennäköisemmin työttömänä hetkellä t, kun tutkia niitä mekanismeja, jotka tuottavat havaittavan yhteyden. Kyseinen yhteys saattaa syntyä siten, että jouduttuaan työttömäksi, työttömän puolison työttömyystodennäköisyys lisääntyy, tai siten, että puolison työttömyys vähentää jo työttömänä olevien työllistymistodennäköisyyttä (pitkittää työttömyyttä). Kyseinen erottelu on hyödyllinen, jotta ymmärtäisimme havaittuun yhteyteen vaikuttavia tekijöitä ja niiden kausaalista voimaa. On täysin mahdollista, että kyseisen yhteyden taustalla on useita mekanismeja (esim. de Graaf – Ultee 2000). On myös mahdollista, että nämä mekanismit toimivat eri aikaisesti, toiset lisäävät työttömäksi joutuneen puolison työttömyystodennäköisyyttä, kun taas toiset vähentävät työttömän henkilön työttömän puolison työllistymistodennäköisyyttä. McGinnityn keskeisin kiinnostuksen kohde oli yhden mekanismin, sosiaaliturvajärjestelmän luomien taloudellisten kannustimien vaikutus työttömien miesten vaimojen työllistymiseen. Tutkimuksessaan McGinnity havaitsi, että miehen työmarkkina-asemalla ei ollut vaikutuksia vaimon joutumiseen työttömäksi. Miehen työttömyys sen sijaan hidasti vaimon työllistymistä Britanniassa, muttei Saksassa. Muiden tekijöiden lisäksi McGinnityn johtopäätöksenä oli, että Britannian vahvemmin tarveharkintaan tukeutuva työttömyysturvajärjestelmä luo työttömien miesten vaimoille 100 kannustinongelmia, jotka vaikuttavat heidän työllistymiseensä. Parhaana todisteena tästä on se, että ko. naisten työllistymisen todennäköisyys putosi rajusti juuri tarveharkinnan tullessa voimaan työttömyyden jatkumisen seurauksena. (McGinnity 2002.) Kyseinen mekanismi vaikutti siis työttömyydestä poistumisen todennäköisyyteen, muttei työttömäksi joutumiseen, ja täten työttömyyden kasautumisen kannalta työttömyyden pitkittymisen vaikutus oli keskeisempi kuin työttömäksi joutumisen vaikutus. On selvää, että näin tarkan tuloksen saaminen poikkileikkausaineistolla olisi ollut mahdotonta. 5.5 Lopuksi Aikaan ja ajallisuuteen liittyvien kysymysten eksplisiittiselle huomioimiselle sosiologisissa teorioissa ja empiirisissä tutkimuksissa voidaan löytää useita perusteita. Vähäisin näistä ei ole kausaalisten selitysten parantaminen. Vaikka sosiologit eivät usein mielellään puhu kausaatiosta, tehdään useissa sosiologisissa teorioissa ja empiirisissä tutkimuksissa oletuksia ja väitteitä eri tekijöiden ja tapahtumien syy-seuraus –suhteista. Sosiologiaa on kritisoitu ajattomaksi tieteeksi (Abbott 2001) – sosiologit eivät riittävästi huomioi aikaan liittyviä kysymyksiä. Tällöin sosiologisesti kiinnostavien tekijöiden vaikutusyhteyksien analysointi kärsii. Tässä artikkelissa olen käsitellyt aikaan liittyvien kysymysten huomioimista sosiologisissa tutkimuskäytännöissä. Erityisen mielenkiinnon kohteena on ollut vertaileva makro- ja mikrososiologinen tutkimus. Aikaan liittyvien kysymysten huomioiminen myös vertailevassa tutkimuksessa avaa parempia mahdollisuuksia yhteiskunnallisten järjestelmien toiminnan ja järjestelmien vaikutusten ymmärtämiseen ja selittämiseen. Vaikka paperini perustuu pitkälti kvantitatiiviselle analyysille niin terminologialtaan kuin esimerkeiltään, niin uskon, että ”logiikaltaan” samoja argumentteja voidaan käyttää hyödyksi myös kvalitatiivisesti orientoituneissa tutkimuksissa (kuten on jo tehtykin). Vaikka tämän paperin keskiössä oli aikaan liittyvien kysymysten käsittely empiirisen tutkimuksen metodologisissa käytännöissä, tulisi samoihin kysymyksiin kiinnittää huomiota myös sosiologisissa teorioissa. Sosiologisissa teorioissa ei juurikaan kiinnitetä eksplisiittistä huomiota aikaan, ja etenkään mekanismien kehittymisen funktionaaliseen muotoon. Ajan huomiointi myös teorioissa voisi lähentää teoriaa ja ajan eksplisiittisesti huomioivaa empiiristä tutkimusta keskenään. Se toisi sosiologisia tutkimusteorioita - sosiaalisten mekanismien puolestapuhujien toivomuksen mukaisesti – konkreettisemmalle ja testattavammalle tasolle. Pitkittäisaineistojen ja niihin sopivien analyysimenetelmien lisääntynyt saatavuus tekee tällaisten teorioiden 101 empiirisen testaamisen mahdolliseksi yhä useammassa tapauksessa. Näillä tekijöillä saattaa olla merkittäviä vaikutuksia sosiologisten teorioiden ja tutkimusten kannalta. Kiinnittämällä huomiota aikaan liittyviin kysymyksiin voidaan vähentää mustan laatikon ongelmaa paitsi vertailevassa tutkimuksessa, niin sosiologiassa yleisemminkin. Tapahtumilla ja prosesseilla on aina ajallinen järjestyksensä, ja selittävät tekijät vaikuttavat selitettävien toteutumisen todennäköisyyteen ajan funktiona. Huolellisesti tehdyt, prosessien dynaamisen luonteen huomioivat hypoteesit auttavat teorioiden selitysvoiman testaamisessa, ja auttavat lisäämään ymmärtämystämme yhteiskunnallisista prosesseista. 102 Kirjallisuus Abbott, Andrew (2001) Time Matters. On Theory and Method. University of Chicago Press: Chicago. Bartolini, Stefano (1993) On time and comparative research. Journal of Theoretical Politics, Vol. 5, No. 2, 131–167. Berk-Schlosser, Dirk – Gisele De Meur (1994) Conditions of democracy in interwar Europe: a Boolean test of major hypotheses. Comparative Politics, Vol. 26, No. 3, 253–279. Blossfeld, Hans-Peter (1996) Macro-sociology, rational choice theory, and time. A theoretical perspective on the empirical analysis of social processes. European Sociological Review, Vol. 12, No. 2, 181– 206. Blossfeld, Hans-Peter – Johannes Huinink (1991) Human capital investments or norms of role transition? How women’s schooling and career affect the process of family formation. American Journal of Sociology, Vol. 97, No. 1, 143–168. Blossfeld, Hans-Peter – Prein, Gerald (toim.) (1998) Rational Choice Theory and Large-scale Data Analysis. Westview Press: Boulder. Blossfeld, Hans-Peter – Götz Rohwer (2001) Techniques of Event History Analysis. New Approaches to Causal Analysis. Second Edition. Lawrence Erlbaum: Mahwah (NJ). Boudon, Raymond (2002) Sociology that really matters. European Academy of Sociology, first annual lecture, 26 October 2001, Swedish Cultural Center. European Sociological Review, Vol. 18, No. 3, 371–378. Coleman, James S. (1990) Foundations of Social Theory. Harvard University Press: Cambridge (Mass.). De Graaf, Paul M. – Wout C. Ultee (2000) United in employment and unemployment? Teoksessa: Welfare Regimes and the Experience of Unemployment, toim. Duncan Gallie ja Serge Paugam. Oxford University Press: Oxford. Dex, Shirley – Gustafsson, Siv – Smith, Nina – Callan, Tim (1995) Cross- national comparisons of the labour force participation of women married to unemployed men. Oxford Economic Papers, Vol. 47, No. 5, 611–635. DiPrete, Thomas A. – John D. Forristal (1994) Multilevel models: methods and substance. Annual Review of Sociology, Vol. 20, 331–357. 103 Esping-Andersen, Gøsta (1990) The Three Worlds of Welfare Capitalism. Princeton University Press: Princeton. Esping-Andersen, Gøsta – Adam Przeworski (2002) Quantitative Cross- national research methods. International Encyclopedia of Social and Behavioural Sciences, 12 649–12 655. Elsevier: Amsterdam. Esser, Hermut (1996) What’s wrong with ‘variable sociology’? European Sociological Review, Vol. 12, No. 2, 159–166. Esser, Hermut (1998) Why are bridge hypotheses necessary? Teoksessa: Rational Choice Theory and Large-scale Data Analysis, toim. Hans-Peter Blossfeld ja Gerald Prein. Westview Press: Boulder. Goldthorpe, John H. (1997) Current issues in comparative macrosociology: a debate on methodological issues. Teoksessa: Comparative Social Research. Methodological Issues in Comparative Social Science, 16, toim. Lars Mjøset, Fredrik Engelstad, Grete Brochmann, Ragnvald Kalleberg ja Arnauld Leira, 1–26. JAI Press: Greenwich. Goldthorpe, John H. (2000) On Sociology. Numbers, Narratives, and the Integration of Research and Theory. Oxford University Press: Oxford. Hedström, Peter – Richard Swedberg (1996) Social Mechanisms. Acta Sociologica, Vol. 39, No. 3, 281–309. Henkens, Kene – Kraaykamp, Gerbert– Siegers, Jacques (1993) Married couples and their labour market status. A study of the relationship between the labour market status of partners. European Sociological Review, Vol. 9, No. 1, 67–78. Holland, Paul (1986) Statistics and causal inference. Journal of the American Statistical Association, Vol. 81, No. 396, 945–960. Kangas, Olli (2003) Institutions and timing: the impacts of legislation on the timing of social insurance. Esitelmä sosiologipäivillä Helsingissä Vertailevan tutkimuksen -työryhmässä 28.3.2003. Kohn, Melvin L. (1989) Introduction. Teoksessa: Cross-National Research in Sociology, toim. Melvin L. Kohn, 17–31. Sage: Newbury Park. LaLonde, Robert (1986) Evaluating the econometric evaluations of training programs with experimental data. American Economic Review, Vol. 76, No. 4, 604–620. McGinnity, Frances (2002) The labour force participation of the wives of unemployed men. Comparing Britain and Germany using longitudinal data. European Sociological Review, Vol. 18, No. 4, 473–488. Mills, C. Wright (1950) The Sociological Imagination. Penguin: Harmondsworth. 104 Mjøset, Lars – Engelstad, Fredrik – Brochmann, Grete – Kalleberg, Ragnvald – Leira, Arnauld (toim.) Comparative Social Research. Methodological Issues in Comparative Social Science, 16. JAI Press: Greenwich. Oppenheimer, Valerie A. (1997) Women’s employment and the gain to marriage: the specialization and trading model. Annual Review of Sociology, Vol. 23, 431–453. Przeworski, Adam – Teune, Henry (1970) Logic of Comparative Social Inquiry. John Wiley & Sons: New York. Ragin, Charles C. (1987) The Comparative Method. Moving Beyond Qualitative and Quantitative Strategies. University of California Press: Berkeley. Ragin, Charles C. (2000) Fuzzy-Set Social Science. Chicago University Press: Chicago. Rueschemayer, Dietrich – Stephens, John D. (1997) Comparing historical sequences – a powerful tool for causal analysis. Teoksessa: Comparative Social Research. Methodological Issues in Comparative Social Science 16, toim. Lars Mjøset, Fredrik Engelstad, Grete Brochmann, Ragnvald Kalleberg ja Arnauld Leira, 55–72. JAI Press: Greenwich. Ruonavaara, Hannu (1995) Teoria ja historia – keskustelua historiallisen Sosiologian dilemmasta. Sosiologia, Vol. 32, No. 4, 253–262. Skocpol, Theda (1979) States and Social Revolutions. Cambridge University Press: Cambridge. Skocpol, Theda (toim.) (1984) Vision and Method in Historical Sociology. Cambridge University Press: Cambridge. Snijders, Tom A.B. – Bosker, Roal J. (1999) Multilevel Analysis: An Introduction to Basic and Advanced Multilevel Modeling. Sage: London. Toivonen, Timo (1999) Empiirinen sosiaalitutkimus. Filosofia ja metodologia. WSOY: Porvoo. Tuma, Nancy B. – Hannan, Michael (1984) Social Dynamics. Models and Methods. Academic Press: New York. 105 6 PERSPECTIVES ON POSTMODERN TOURISM AND AUTHENTICITY Pekka Mustonen Abstract Tourism cannot be seen only from one perspective. The phenomenon contains premodern, modern and postmodern features, of which the latest have been widely discussed during the recent two decades. The aim of this paper is to evaluate critically the most common definitions of postmodern tourism in addition to McDonaldization thesis, which by its nature is a totally adverse phenomenon. Both McDonaldization and postmodern features can be found in contemporary tourism, and neither of them shows the whole picture of tourist culture of today. Nevertheless in tourism research, there is a clear role for these “fundamental” theories. They could give information on different sides of the complex phenomenon. Also the role of authenticity has played an essential part in discussions on tourism research. In this paper it is suggested that the MacCannell’s original idea of tourists searching for authentic experiences, could be improved. The discussion on authenticity should concern tourists’ own views and opinions and thus it could be asked if authenticity actually is a dynamic concept? From this perspective even the most McDonaldized destinations and ways to travel could be authentic. Tourists can be seen to be searching for authentic tourist culture and touristic experiences, which not necessarily have anything to do with old and historical and this way authentic traditions and sights. Thus every kind of tourism can offer deep and authentic experiences for tourists. KEYWORDS: McDonaldization, postmodern tourism, authenticity 6.1 Introduction The role of authenticity has played an essential part in discussions on tourism research. Dean MacCannell (1973; 1999) is assuredly the most famous theorist, when this complex concept is concerned. According to him tourists 106 searching for authenticity actually find staged authentic sights, which have been separated from their original context. Now along with discussions on postmodern tourism, it is necessary to evaluate the concept in relation to postmodern theories. This is one aim of this paper. Other is to evaluate the attributes of post-tourists presented first by Feifer (1985) together with Ritzer’s (e.g. 1998) McDonaldization thesis, which is in a way theoretically an adverse phenomenon to postmodernism. Also other aspects of postmodern tourism are examined, but the paper will mainly concentrate on Feifer’s classic notion. Both McDonaldization and postmodern features could be found in contemporary tourism, and neither of them shows the whole picture of contemporary tourism culture. Tourism can not be seen only from once perspective. This leads to the main question of this paper: Is it possible to draw a line between modern and postmodern tourism and tourist? The answer is quite evident. Even though it might be possible, it is not necessary. More important is to examine whether these postmodern ideas cast new light on tourism (see Ritzer – Liska 1997, 97). 6.2 Postmodern tourism and tourists Newer trends in tourism and consumption in general have brought postmodernism into discussion and therefore the way to see tourism as a contemporary phenomenon has been changing.10 Tourist has even been considered as a metaphor for postmodern man (Bauman 1993, Featherstone 1995, Jokinen – Veijola 1997). Even so, contemporary tourism has traditionally been considered as a modern phenomenon (Cohen 1995; Uriely 1997). Together with assumed change from modern to postmodern, also globalization has been widely discussed. Though globalization as it is usually understood, could be seen to have begun in the middle of the 20th century (Anttiroiko 2003), its most common manifestations, e.g. influence and power of multinational companies, have been tremendously increasing in the 70’s and 80’s. Thus both postmodern and globalization appeared into wider sociological discussion about in the same time. Coincident or not, there are connections between postmodern and globalization. These two theoretically different notions are just two different angles to see the contemporary world. Neither gives enough information about current changes and changes of the 10 For a brief review of theories of modern and postmodern tourism, see Uriely (1997). 107 past, but together they might enlighten factors behind current changes in social life. Often when globalization is concerned, example of a fast food chain McDonald’s has been raised to describe the phenomenon. Multinational fast food brands could be seen as good examples of extremely modern way to supply goods to consumers. Fordist methods and assembly lines are traditionally considered as signs of efficiency, which could be seen to be an essential feature of modern. Methods that large multinational companies and chains like McDonald’s are using could be transferred to the other parts of economy like tourism as well (see Ritzer 1998). Wood (1997, 2) states that even more than McDonald’s, international tourism symbolizes globalization through its linkage of economic, political and sociocultural elements (see also Kosonen 1999 for discussion on the concept of globalization). Quite often almost every current change in societies are tried to be explained by globalization, which as well as McDonaldization thesis, should be considered just as one point of view and attempt to explain modernization and contemporary consumption culture. In addition to restaurants, shopping malls, credit cards and even universities, Ritzer (1998) connects his theory of McDonaldization also with tourism. According to him, McDonaldization is a paradigm of rationality and it could be thought as a completely modern approach to tourism, because more and more touristic experiences tend to be efficient, calculable, predictable and non-surprising. Ritzer’s most common example is Disney, especially Disney theme parks, which can be seen as the most perfect models of post-tourism (Bryman 1995, 178–179). Ritzer states that tourism in general has been McDisneyized11 to some extent, and that tourists want their experiences to be as McDonaldized as their day-to-day life. He claims that even the most specialized types of tourism will apply principles of McDonaldization, and the whole process is made without it seeming to be McDonaldized. Taken to the extreme, McDonaldization undermines the fundamental reason to tour, because fully McDonaldized destinations do not offer new or different experiences, or experiences of change, which are thought to be essential for tourists (see Smith 1989). Diversification of tourism is a trend, which is commonly recognized and agreed. In spite of that, Ritzer (1998, 137) is not willing to give up his thesis of McDonaldization. He states that because contemporary lifestyle in western countries and destinations is so similar, it is not necessary to standardize the product itself. It is not surprising, that he advises to take standardized meals 11 Disney (McDisneyization) is probably the most commonly used example of McDonaldization in tourism. (See Bryman 1995) 108 for an example. According to Ritzer, similar chains also in other sectors of economy make tourist sites familiar and comfortable for majority of tourists. So despite the fact that the freedom of choice in postmodern West has increased and made it possible to choose also lifestyle (usually defined by consumption) according to one’s taste (Miles 1998), people still look for standardized ways to consume. There are also other restrictions, which diminish the impact of postmodern. The main idea behind postmodern consumption patterns is a diminished effect of modern structures (e.g. social class). Connections between social structures and consumption might have to some extend become weaker (Featherstone 1991; Toivonen 1991 & 1997; Urry 1995; Miles 1998), but for example monetary restraints still create limits to consumption, and thus consumers’ choice of alternatives is inevitably imperfect (see Räsänen 2000). Structures have not lost all their importance as a factor for influencing people’s behaviour. In addition to income, these structural factors include for example socio-economic status, type of household and place of residence (see Räsänen 2003). Postmodern tourists, or post-tourists, could be seen as opposition to mass tourists. Post-tourists attempt to gain access to some level of authentic experience by venturing away from mass tourist routes (Black 2000, 254; compare also to Munt 1994). It was Feifer (1985), who first brought the concept of post-tourists into discussions. Her idea of post-tourist is the most commonly used and practically all discussions on post-tourists have this original notion as a background. The concept of post-tourists has been taken further and developed by Urry (1990), and later by for example Rojek (1993), Munt (1994) and Ritzer (1998). Post-tourists are said to be aware of their role of being tourists in the world of simulations (see Baudrillard on simulations, e.g. 1996), where authenticity (MacCannell 1999) has become scarcity. When modern tourists were seriously searching for authenticity, post-modern tourists are in a playful search for enjoyment (Cohen 1995, 21). They can also gaze (Urry 1990) typical tourist attractions at home. In addition to these two main characteristics, presented by Feifer (1985), post-tourists have a lot of alternatives and different types of travelling can be combined. Rojek (1993) presents a couple of attributes more. First of all, post-tourist accepts the commodification of tourism. She/he realizes how all the products connected to tourism are just manifestations of consumerism. This first attribute can be derived also from Feifer’s idea of conscious post-tourists. Second considers tourism as a separate form of behaviour, which is performed without any greater goal, for its own sake. Third attribute that Rojek adds to be typical to post-tourist is a consumption of signs (see also Miles 1998, 23). Despite being characteristic to postmodern 109 culture, signs have been considered to be important also for modern tourists (compare to McDonaldization of tourism, Ritzer 1998). In addition to these features mentioned above, Munt (1994) connects postmodern tourism, especially its associations with Otherness with the new social movements and environmentalism. By utilizing the notion of Otherness, Munt wants to create a gap between “real-holiday travellers” and “mass- packaged tourists”. Post-tourists for Munt are mainly middle-class people to whom oppositional travel is a cultural asset (see also Black 2000, 255) and who want to make a clear distinction between themselves and traditional, modern, mass tourists. In addition to underlining the individualism, Munt states that also tour operators have changed and are now capable to offer more flexible and customized holidays. In addition to postmodern tourists, the “postmodern change” could also be found in the destinations. According to Sardar (1998, 134), tourists look at Other societies as locations for culture, which in turn has disappeared in the West. For example, classic examples of postmodernism in tourist destinations, theme parks, are based on simulated experiences (Baudrillard 1988, MacCannell 1999, Ritzer 1998, Munt 1994). On the other hand, Ritzer (1998) uses theme parks as examples for his McDonaldization thesis of tourism, which could be seen to describe the most extreme form of modern and standardized tourism. Basically these two ways to see theme parks do not differ much from each other. Theme parks could be considered postmodern because of simulated experience they provide, and modern because of the standardization. As technological change and emerging globalization led to the rise of modern tourism, probably these same phenomena, just taken further, have also created post-tourists? Like theories of post-tourists state, newest technological innovations have made it possible to gaze sights without really moving. Anyway, there are fundamental differences between real tourism, which must include physical movement12 (see Selänniemi 1996, see also discussion on the topic in Cohen 1995), and virtual tourism. In virtual reality, it is possible to adopt different roles and “do” things, which are not possible without new technological devices. In virtual reality, material constraints of real life do not apply. Virtual reality absorbed for example into Internet Related Chat (IRC) 12 Official definitions usually created for statistical purposes are sometimes not close enough to real life, but here the definition of WTO (World Tourism Organization) must be mentioned. According to WTO tourists is "any person who travels to a country other than that in which she/he has her/his usual residence but outside her/his usual environment for a period not exceeding 12 months and whose main purpose of visit is other than the exercise of an activity remunerated from within the country visited". (http://www.world-tourism.org/market_research/facts&figures/menu.htm 4.11.2003) 110 clearly precedes virtual tourism. Probably the role of virtual tourism as a fairly abstract idea will always remain (compare to Poster 1996). Anyhow, Matala (2003) considers virtual tourism as “pre-tourism”, which makes travelling easier and creates opportunities, also for the supply side. On the other hand, she states that technological innovations, which bring virtual tourism closer to reality, could help tourists to recover from finished tour. The same could also happen before the trip and like this, technological innovations like the Internet and other forms of media have increased tourists’ knowledge of the nature of tourism and of possibilities, and thus enabled more diversified, “postmodern” tourism. Even though this is evidently true, it could be stated that this kind of postmodernity is only due to technical development. So from this point of view, it is quite absurd to try to find some fundamental endogenous differences between past and present and it might sound irrelevant to think of a totally new species of tourists, e.g. post-tourists, because also bygone tourists might have been acting similarly if they just had possibilities. (See Ahtola 2002 and Kostiainen 2003 for discussion on globalization and tourism) So considering this discussion, it could not be stated that there are post-tourists, who differ mentally from other tourists or tourists from the past. 6.3 Some alternative thoughts There are several evident contradictions between ideas of McDonaldization and postmodernity of tourism. Sometimes these both have even been explained by utilizing the same arguments. For example, virtual tourism is said to represent postmodern tourism whilst Ritzer (1998, 140) states that virtual tourism fits well to the picture of McDonaldization. After all, tourism as an economic and social phenomenon was once traditional (or premodern if derivations from the word ‘modern’ are used), then became modern and is now probably turning into something new, say postmodern. Same person could represent premodern, modern and postmodern features of tourism even during the same trip. Thus when tourism is concerned, the whole idea of a clear postmodern era must be questioned. Instead, it could be stated that there is just a complex phenomenon called tourism, which has traditional, modern and postmodern features. There are many examples that could be presented here. One quite apposite is a phenomenon, which was born on the basis of people’s urge to practise yoga in an authentic atmosphere, in India, where the whole philosophy was born. Every year dozens of tourists from Finland visit India for that purpose. In one sense this kind of tourism is very close to pilgrimage. On the other hand, yoga-tourists take advantage of 111 modern facilities; they live in guesthouses, eat in restaurants, read their e-mails and travel by train and plane. They have most likely searched for information by utilizing the newest technology, and in addition to that, they most likely are able to throw away their regular self and be real and honest, playful tourists. Many aspects of the contemporary society, which are said to represent postmodernity, could also be found from tourism, and vice versa. As Ritzer (1998, 132) states, new means of consumption are not modern or postmodern but social phenomena, which can be seen from the both perspectives. These new ways to consume, that have brought postmodernity into discussion have, to some extent, existed also before. Nor are environmentally aware, individual and initiative tourists only phenomena of today. And although access to the sources of information is easier now than a century ago, there must have been always tourists who were aware of their roles. But in the same time the amount of alternatives is increasing, and “playing postmodern game” during tours, which could be defined as modern or even premodern, is now possible. Thus postmodern features may sometimes even strengthen modern ways of tourism, like traditional mass tourism. Actually Selänniemi (1996, 170-172) criticizes the paradoxal idea of post-tourists by asking if finally all tourists who realize being tourists, and that tourism is a form of business, are post-tourists? The fact that the classic definitions of post- tourists are so broad actually diminishes their ability to explain social phenomena. As stated earlier, tourism by its nature is a modern phenomenon, which has been changing tremendously and some of these changes have been called postmodern. In addition to this there are also traditional, premodern characteristics, which have slightly diffused to newer tourism trends. For example behaviour that is common to pilgrims, who probably were one of the very first tourists (see Ilola 1994), could be found in many forms of contemporary tourism. Modern mass tourism has even been compared to modern pilgrimage (Ilola 1994; Selänniemi 1996). The number of examples how these kinds of connections could be found is endless: Someone may visit ancient sites or go to see famous paintings. This rendezvous could actually be the main motive of the trip. Someone, on the other hand, may want to go to visit a brewery where her/his favourite beer is brewed. For these “pilgrims” experience or reward could be compared to religious experience. Like a holy place is sacred for the devoted pilgrim, for the tourist different kinds of touristic experiences could be sacred (compare to Selänniemi 1996, 181–183). Sometimes even more, because for some pilgrims, the reason for travelling to Holy Places might not be in a sense religious but just a way to higher one’s status (see Ilola 1994). So like it was in the middle of the 80’s, it is still too 112 early to state that the clear leap from modern tourism to postmodern tourism has occurred. So despite the enormous changes, modern features will most likely remain also in the future. If the most commonly used characteristics of postmodern tourism (e.g. Feifer 1985; Urry 1990) are kept in mind, modern tourism to the South (Selänniemi 1996) contains surely postmodern features from tourists’ point of view. If tourists realize their role as tourists in inauthentic destinations, (see Feifer 1985, see also MacCannell 1999 for discussion on authenticity), even the most “McDonaldized” forms of tourism, like Disney-Parks (Bryman 1995, 177), could be seen as postmodern in that sense. As Rojek (1997, 62) states, post-tourists may voluntarily and ironically play the role of mass tourist, because it is not possible to avoid the influence of signs (see MacCannell 1999), on which tourism is often based. But here again, a clear line between post-tourists and ordinary mass tourists is difficult, probably impossible to draw. One reason for this is that mass tourists are heterogeneous group of individuals (Selänniemi 1996, 230). Thus it can be stated that all mass tourists are post-tourists, and for them it is possible to play the role of mass tourist, because they in a way are mass tourists. This “ironic” game (compare to ibid., 199-200) of tourists occurs in the liminoid (ibid.; Graburn 1989), where tourists’ behaviour may differ significantly from their behaviour at home. Here ideas of post-tourists would state that the leap into this state of liminality must to some extent be intentional. Other way to see mass tourism as a postmodern phenomenon is to think of all those possibilities that mass destinations offer behind the advertisements and scenes. Tourists may create their holidays and trips according to their own taste. They may stay the whole week or two by the pool, but they may also join in guided tours, because they think that it might be a good and harmless way to see sights, not because they could not do the same tours by themselves (see Rojek 1997, 62). During rest of the holiday they might make their own individual tours around the destination. Thus individualism in tourism must be understood to be wider phenomenon than for example backpacking or initiative booking of tickets, although the most important feature of individual travel according to Hyde&Lawson (2003) is a lack of planning and prebooking of vacation elements. Thus as one main characteristic of the postmodern is the rise of individualism, it is natural that many features of postmodern can be found from individual tourism.13 Sometimes these features have been said to be a response to modern mass tourism. But unlike Urry (1990) claims, modern package tour remains alive (Ritzer 1998, 137) and is even increasing (see Honkanen 2002). This fact does not necessarily mean that 13 For discussion on New Tourism, see Poon (1993). 113 the idea of postmodern tourism should be rejected. Instead, this increasing popularity of traditional mass tourism could even be the way how postmodernism appears today in tourism. Recently, some might say through postmodern change, traditional mass tourism has become more flexible and freedom of choice has increased. Though it must not be forgotten that in the same time when also traditional tourism has reached more individual forms than before, touring to non- standardized destinations is still possible without using any standardized services of standardized companies. And despite McDonaldization, the trend contrary to postmodernization, people may act more individually than it is usually believed (Honkanen 2002, 13). Thus it can be dangerous to try to transfer the idea originating from fast food to everything. These attempts do eventually diminish McDonaldization thesis’ ability to explain social phenomena. This can also be connected with theories of postmodern tourism, which have been criticized to be ethnocentric or even westernised. Postmodern theories often consider social phenomena from western perspective, although tourist destinations and sights can be consumed and interpreted in countless different ways, which may differ from those stated by western postmodern theories (see Edensor 1998, 1–9). 6.4 Post-tourists searching for authenticity? The idea of authenticity (see MacCannell 1973; 1999) is essential when post- tourists and post-tourism are discussed. There is not any rule stating what is authentic and what is not. Both authenticity and the concept of reality, which join easily together, are difficult to interpret. It could be asked what is the true real identity of the object and how it is possible to find? For Baudrillard there are just simulations, no real or unreal. So authenticity or reality must depend on the point of view. It depends on the person who is gazing the target. Same destination can be authentic to someone and inauthentic to someone else. Like all the tourists can be considered as post-tourists by using some of the several arguments, also everything is authentic to someone. Post-tourist may travel to the some traditional mass tourism destination, and realize that the destination is authentic in a touristic way. The destination does not claim to be anything more than just a mass tourism destination. For someone, this could be a good reason to go to destination. Someone might even want to travel for example to Playa Del Ingles (see Selänniemi 1996 for discussion on package tours) because it is a classic example of a mass tourism destination totally created for tourists and truly authentic in that way. 114 Classic theories of modern tourism are somewhat valid also in postmodern world (see Uriely 1997, 983). Also post-modern tourists can be searching for authenticity (Munt 1994, Urry 1990). In addition to pure simulations, there are artificial tourist destinations where experiences can be stated as inauthentic, but still there are also tourists who seek for new, real experiences. Anyway, as stated earlier, simulations or experiences, which might seem inauthentic, can be meaningful for someone. It must be remembered that even before the postmodernity of tourism was a topic for the discussion, Cohen (1979, 180) claimed that tourists might seek different kinds of experiences. Ritzer (1998, 146) indirectly refers to this same idea when he says that the people who are living in the postmodern world dominated by simulations want simulated experiences also when they are travelling. Ritzer’s idea is that tourists look for these similarities because they are used to it, and they feel more confident when surrounded by something familiar. This differs to some extent from post-tourists’ way to see these simulations. According to presumptions of post-tourists’ characteristics, post-tourists should be aware of these simulations, and they may look for them intentionally. And even if they were not looking for them, they would enjoy their holiday despite these simulations. They understand that existence of simulated experiences is necessary and unavoidable part of tourism today. Contrary to this, Ritzer (1998, 146) states that tourist could not even recognize authentic sight even if they met one. And because “real” has disappeared in the world of simulations, and as simulating continues, consumers finally meet simulations of simulations (ibid., 147; see also Baudrillard 1996). Finally Ritzer (1998, 145) states that if tourists are seeking for authenticity, and as the postmodern society is increasingly dominated by simulations tourists are doomed in their search. (See also MacCannell 1999) He continues contrary to MacCannell that tourists are today searching for inauthenticity. To avoid serious contradictions with postmodern theories, Ritzer’s statement could be justified only in the case of a clear definition of authenticity. Because such a definition does not exist, and if postmodern theories of tourism are still kept in mind, the concept of authenticity must be seen as dynamic. As culture is a dynamic concept and thus changing all the time, also sense of authenticity must change. Staged authenticity becomes authentic. On the other hand, if all (or at least some of) that used to be authentic in chronological sense (the older the more authentic) has disappeared, is there any reason to use the concept anymore? By adjusting the idea of authenticity to be more in conjunction with postmodern theories and contemporary world, MacCannell’s idea of tourist seeking authenticity could still be useful. The idea could still create understanding of fundamental reasons to tour. Thus when postmodern tourism 115 is concerned, questioning the importance of the concept of authenticity is not necessary. After all, it must not be forgotten that postmodern diversification of tourism leads to situation where it is not possible to state who is searching authenticity (or disauthenticity) and who is not, because post-tourists are able to change their roles during the holiday (compare to Selänniemi 1996) and even when not (yet/anymore) on holiday. 6.5 The challenge of sustainability As stated earlier, postmodern consumers are to some extent capable to choose their lifestyle, manifested usually by consumption (Räsänen 2000). Postmodern consumer according to one view (e.g. Munt 1994) wants to separate herself/himself from others, and thus to gain satisfaction. Munt also adds environmentalism in the middle of the matrix of post-tourism. This is possible if postmodernism is extended to include also postmaterialism, and if Feifer’s and Urry’s ideas of post tourists are concerned, this probably should be done. One of those main features of post tourists was tourists’ ability to consciously choose the way of travelling and thus the honest interest to non- material issues must be added to the extensive net of postmodern tourism. The difference between postmaterialism and postmodernism can be presented by using a simple matrix: Postmodern thinking concentrates on the present and postmaterialism on the future. Further, postmodern values are heterogeneous whilst postmaterial values are somewhat similar. So according to this distinction, when postmodern tourists concentrate on their present welfare, they may for example attend to “eco-tour” only because her/his social status requires so14. They might pretend to be environmentally friendly or socially aware just to raise their status. So from this point of view environmentalism is surely a value, which could be adopted even without real intention. Postmaterial behaviour instead must look further, because idea of giving up material welfare does not easily fit into the picture of rational modern consumer maximising her/his welfare. Although diverse postmodern values also include environmentalism, they are easier to connect with postmaterialism, (see Mustonen 2003, 36-37) and environmentally, socially and culturally aware tourists are representing mainly postmaterial values. Recent changes, postmodern or not, have given birth to many new forms of tourism. Some of these can be locally sustainable and can maintain and even improve locals’ welfare and environment. In Finland for example, some farms maintain small-scale tourism services for people who want to stay a few days 14 Compare to discussions on pilgrims. 116 outside of the cities and experience life in the countryside. Experience itself might be far from the normal farm life, but again it is not possible to state that it is inauthentic for tourists. Anyhow, this kind of tourism can be locally ecologically, economically and socially sustainable, partly because the biggest threat to sustainability, too large tourists masses, is usually not a problem in Finland when this kind of tourism is concerned. Other examples of new forms of tourism, which can be locally sustainable, consist for example of volunteer tourism, small scale and well-planned eco-tourism and some newer forms of urban tourism. In spite of that some of these new alternatives may be locally sustainable and thus answers to many problems in host communities, an increase in alternatives in general is a clear threat to sustainability when the entire field of global tourism is concerned. Post-tourist might well be honestly interested in sustainability of tourism and may participate in so-called eco-tours. Later same tourist may play conscious role of traditional mass tourist in a Playa del Anywhere (see Selänniemi 2001). Thus due to emergence of new forms of tourism e.g. eco-tourism, and tourists’ possibility to choose their own roles and types of touring, these “postmodern” changes can lead to increasing tourist flows. On the other hand if post-tourist is aware of the role of tourism in destinations, should not she/he also be aware of its impacts? From this question it is not difficult to conclude that if there is not a clear epoch of postmodernism, there is not a clear picture of post-tourist. There hardly are tourists who in the same time play an ironic role of tourist in an “inauthentic” destination, gaze sights virtually, know and take advantage of alternatives and are aware of impacts. The web of impacts is so huge that it is difficult for an individual tourist to be aware of the relations between the dimensions of impacts and sustainability. The selfish side of postmodernism (Mustonen 2003, 37; see also Inglehart 1997, 76–77) brings again modern and postmodern ideas of behaviour closer together. According to classic consumer theories, in modern world, consumers are said to be using all the available information to make economic decisions to maximize their benefits. (Honkapohja 1996) Their expectations are said to be rational which means that no systematic mistakes are made. Although the hypothesis of the rational consumer is theoretical, the idea can be connected to post-tourists’ sense of knowledge. Although rationality to many does not include knowledge or even interest to sustainability issues, consumers may include these issues to their utility basket. The main idea behind post- materialism and environmentalism is a belief that an act for the environment or for the others will finally have an effect to one’s own life. (Mustonen 2003, 37; Brand 1999, 645–646) So post-materialists and thus environmentalists think that rational behaviour would finally lead to sustainable behaviour. The 117 same idea could also be transferred to enterprises. In the long run, sustainable use of resources, gives the biggest benefits. From this could be derived that economic behaviour in postmodern times could be described in the same way than in modern times - through the hypothesis of utility maximization. 6.6 Discussion Ideas of post-tourists (e.g. Feifer 1985) and McDonaldization of society (e.g. Ritzer 1999) are totally contrary to each other. Postmodern theories of tourism, which include the notion of post-tourists, state that modern structures are disappearing or even have disappeared, and that post-tourists are aware of these changes and are able to take advantage on them. McDonaldization thesis instead claims that standardization of destinations and ways to travel have gone so far that people on holiday have started to look for same kind of circumstances that they could find at home. Taken to the extreme, McDonaldization thesis even questions the fundamental reason to tour. But despite evident diversification of tourism, Ritzer (1999) states that McDonaldization thesis is still holding strong. It is clear that both of these trends are visible in contemporary tourism. Nevertheless, there is a clear role for these “fundamental” theories in tourism research. They can give information on different sides of the complex phenomenon, tourism. Nevertheless, McDonaldization thesis somewhat ignores the fact that tourism as a phenomenon has become more and more complex and also tourists’ awareness about themselves and the phenomenon has increased. Also ideas of post-tourists partly fail because they do not pay attention to the other sides of the coin. Ideas of postmodern tourism cannot be generalized, because old structures still have a strong effect on peoples’ behaviour. There are also forms of travel like pilgrimage, which do not fit easily into postmodern pattern. On the other hand, according to postmodern consumption theories, essential to contemporary consumption is consumers’ own desires and increasing possibilities to fulfil them. So if thoughts about postmodern are like this taken further to concern almost everything, there is a danger that its ability to explain social phenomena diminishes. The paradigm of authenticity (see MacCannell 1973; 1999) is essential when modern tourism, but also when postmodern tourism is concerned. Modern tourist is searching for the authentic experiences, but usually fails, because authentic experiences are more and more difficult to find. Postmodern tourist according to some views is instead searching for inauthentic experiences and something that is familiar to her/him (Ritzer 1999). Therefore destinations are turned into suppliers of simulations, even simulations of 118 simulations (Baudrillard 1996). In this article the concept of authenticity was evaluated and conclusion was that if it is remained untouched, there hardly is any good reason to use the concept at all, because “genuinely” or chronologically authentic destinations are increasingly difficult to find. Thus it might be possible to improve MacCannell’s idea and state that tourists are still searching for authentic tourist culture and touristic experiences. Like this even the most McDonaldized destinations and ways to travel, like cruises or tours to mass tourism destinations like Playa del Ingles, could be authentic from touristic angle. Thus every kind of tourism can offer very deep and authentic experiences for tourists. The concept of authenticity and its relevance is not easy to study empirically because the concept can be interpreted by several ways. What is authenticity and is it a dynamic concept? It surely is dynamic especially when individuals are concerned. Thus the discussion is or should be about tourists’ own view and opinions. Deeper investigation of authenticity could be possible by approaching the problem from totally different point of view. Selänniemi (1996) for example presents quotations from tourists’ diaries in his study on Finnish mass tourists. From these short writings it is possible to find postmodern features and tourists searching for escape (see e.g. Rojek 1993) and touristic authenticity – and unlike MacCannell (1999) states, even succeeding in their task. 119 References Ahtola, Janne (2002) Globalisaatio ja Matkailu. In: Dialogus. Historian taito, ed. by Meri Heinonen, Leila Koivunen, Sakari Ollitervo, Heli Paalumäki, Hannu Salmi and Janne Tunturi. Kirja-Aurora: Turku. Anttiroiko, Ari-Veikko. Kunnat Globalisaation aikakaudella. , 4.8.2003. Baudrillard, Jean (1996) Disneyworld Company. In: Liberation -newspaper. Paris. 4.4.1996. , 5.8.2003. Baudrillard, Jean (1988) Simulacra and Simulations. In: Jean Baudrillard, Selected Writings, ed. by Mark Poster. Stanford University Press: Stanford. 166–184. Bauman, Zygmunt (1993) Postmodern Ethics. Blacwell. Oxford. Black, Paula (2000) Sex and travel: making the links. In: Tourism and Sex: Culture, Commerce and Coercion, ed. by Stephen Clift and Simon Carter. Pinter: London and New York. 250–264. Brand, Peter (1999) The Environment and Postmodern Spatial Consciousness: A Sociology of Urban Environment Agendas. Journal of Environmental Planning & Management. Vol. 42, No. 5, 631– 648. Bryman, Allan (1995) Disney and his worlds. Routledge: London Cohen, Erik (1979) A phenomenology of tourist experience. Sociology. Vol.13, No. 2, 179–202. Cohen, Erik (1995) Contemporary tourism – trends and challenges. In: Change in Tourism. People, Places and Processes, ed by. Richard Butler and Douglas Pearche. Routledge: London. Edensor, Tim (1998) Tourists at the Taj. Performance and meaning at a symbolic site. Routledge: London and New York. Featherstone, Mike (1991) Consumer Culture & postmodernism. SAGE Publications: London. Featherstone, Mike (1995) Undoing Culture: Globalization, Postmodernism and Identity. SAGE Publications: London. Feifer, Maxine (1985) Tourism in History: From Imperial Rome to the Present. Stein and Day: New York. Graburn, Nelson (1989) Tourism: The Sacred Journey. In: Hosts and Guests. The Anthropology of Tourism, ed. by Valene L. Smith, 21–36. University of Pennsylvania Press: Philadelphia. 120 Honkanen, Antti (2002) The Postmodern Diversification of Tourism in Finnish Tourists’ Destinations during the 1990s? In: Issues in Modern and Postmodern Tourism, ed. by Pekka Mustonen, 5–18. Discussion and Working Papers No 4. FUNTS: Savonlinna. Honkapohja, Seppo (1996) Odotukset ja oppimisprosessit taloustieteen tutkimuskohteena. Kansantaloudellinen aikakauskirja, Vol. 92, No. 1, 85–91. Hyde, Kenneth F. – Lawson, Rob (2003) The Nature of Independent Travel. Journal of Travel Research, Vol. 42, No. 1, 13–23. Ilola, Heli Katariina (1994) Pyhiinvaellukset ja turismi. Katsaus matkailun kahteen tyyppiin. Pohjois-Suomen tutkimuslaitos. Oulun yliopisto: Oulu. Inglehart, Ronald (1997) Modernization and Postmodernization: Cultural, Economic and Political Change in 43 Societies. Princeton University Press: New Jersey. Jokinen, Eeva - Veijola, Soile (1997) The Disoriented Tourist. In: Touring Cultures, ed. by Chris Rojek and John Urry, 23–51. Routledge: London and New York. Kosonen, Pekka (1999) Globalisaatio sosiologisena käsitteenä. Sosiologia, Vol. 36, No. 3, 181–192. Kostiainen, Auvo (2003) The History of Globalization Related to the History of Tourism. Muuttuva Matkailu 2/2002. IAST 2003 Special Issue. FUNTS. Joensuun yliopistopaino: Joensuu. MacCannell, Dean (1973) Staged Authenticity. The American Journal of Sociology. Vol. 79, No. 3, 589–603. MacCannell, Dean (1999) (1976; 1989) The Tourist: A New Theory of the Leisure Class. University of California Press. Berkeley: Los Angeles and London. Matala, Sari (2003) Virtuaalimatkailua esielämyksiin. Kulttuuritutkimus, Vol. 20, No. 2, 53–65. Miles, Steven (1998) Consumerism as a Way of Life. SAGE: London. Munt, Ian (1994) The ‘Other’ Postmodern Tourism: Culture, Travel and the New Middle Classes. Theory, Culture & Society, Vol. 11, 101– 123. Mustonen, Pekka (2003) Environment as a Criterion for Choosing a Holiday Destination – Arguments and Findings. Tourism Recreation Research. Vol. 28, No. 1, 35–46. Poon, Auliana (1993) Tourism, technology and competitive strategies. CAB: Wallingford. Poster, Mike (1996) Postmodern Virtualities. In: Cyberspace/Cyberbodies/Cyberpunk. Cultures of Technological 121 Embodiment, ed. by Mike Featherstone and Roger Burrows, 79– 95. SAGE Publications: London, Thousand Oaks and New Delhi. Ritzer, George - Liska, Allan (1997) ’McDisneyization’ and ’Post-tourism’: Complementary Perspectives on Contemporary Tourism. In: Touring Cultures. Transformations of Travel and Theory, ed. by Chris Rojek and John Urry, 96–109. Routledge: London and New York. Ritzer, George (1998) The McDonaldization Thesis. Explorations and Extensions. SAGE Publications: London, Thousand Oaks and New Delhi. Rojek, Chris (1993) Ways of Escape. Modern Transformations in Leisure and Travel. Routledge: London. Rojek , Chris (1997) Indexing, Dragging and Social Construction of Tourist Sights. In: Touring Cultures. Transformations of Travel and Theory, ed. by Chris Rojek and John Urry, 52–74. Routledge: London and New York. Räsänen, Pekka (2000) Kulutusvalintojen postmodernit piirteet ja rakenteelliset reunaehdot. (The Postmodern Features and Structural Conditions of Consumers’ Choices). Sosiologia, Vol. 37, No. 3, 228–242. Räsänen, Pekka (2003) In the Twilight of Social Structures: A Mechanism- Based Study of Contemporary Consumer Behaviour. University of Turku: Turku. Sardar, Ziauddin (1998) Postmodernism and the Other: The New Imperialism of Western Culture. Pluto Press: London & Chicago. Selänniemi, Tom (1996) Matka ikuiseen kesään: Kulttuuriantropologinen näkökulma suomalaisten etelänmatkailuun (A Journey to the Eternal Summer: The Anthropology of Finnish Sunlust Tourism). Suomalaisen Kirjallisuuden Seura: Helsinki. Selänniemi, Tom (2001) Pale Skin on Playa del Anywhere: Finnish Tourists in the Liminoid South. In: Hosts and Guests Revisited: Tourism Issues of the 21st Century, ed. by Valene Smith. Cognizant Communications: New York. Smith, Valene (1989) Introduction. In: Hosts and Guests: The Anthropology of Tourism, ed. by Smith, Valene, 1–17. University of Pennsylvania Press: Philadelphia. Toivonen, Timo (1991) The melting away of class differences? Consumption differences between employee groups in Finland 1955-1985. Social Indicators Research. Vol. 25, 151–176. Toivonen, Timo (1997) Food and social class. Journal of Consumer Studies and Home Economics. Vol. 21, No. 4, 329–347. 122 Urry, John (1990) The Tourist Gaze: Leisure and Travels in contemporary Societies. SAGE Publications: London, Newbury Park and New Delhi. Urry, John (1995) Rethinking Class. In: Social Movements and Social Classes: The Future of Collective Action, ed. by Louis Maheu, 169–181. SAGE: London. Uriely, Natan (1997) Theories of Modern and Postmodern Tourism. Annals of Tourism Research, Vol. 24, No. 4, 982–985. Wood, Robert E. (1997) Tourism, Ethnicity & the State in Asian & Pacific Societies. University of Hawaii Press: Honolulu. 123 7 INFORMATION TECHNOLOGY AND CONSUMPTION HABITS AMONG YOUNG PEOPLE FROM RIO DE JANEIRO Sueila Pedrozo Abstract The development of the information technology and media in general have brought not only more information and communication, but also new consumer cultures as well as new kinds of consumer, especially among young people. In Brazil, the information society along with the Internet and the mobile phone is just beginning, since a large segment of the population still has no access to these devices due to factors as education and financial resources. Both this paper and the empirical research are based on the original survey Consumer Cultures of Young People in the Changing Information Society, carried on in 2001 in Finland and replicated in Rio de Janeiro (approximately the same population as in Finland), Brazil and two rural areas, between February and May 2003. The survey focused on the consumption habits and lifestyles of Brazilian teenagers 16-20 years old, and the data were collected from middle-level private schools, federal public schools and first year university courses. The sample included 736 students required to fill a questionnaire (60 questions) covering themes as the individual consumption and money use, mobile phone and the Internet, issues on environment and security. This survey aims to map young people’s attitudes towards IT and consumption in Rio de Janeiro and identify eventual connections between them. A future use of the data might be a cross-cultural study between Finland and Brazil. KEYWORDS: information technology, consumption, young people 7.1 Introduction In a country of more than 175 million people, a GNP per capita of 2,955 € and deep contrasts; income distribution – one of the most perverse, lack of education, unemployment, and child labor (5,4 million children from 5 to 17 124 years of age); the IT explosion is really surprising. In the 90’s cellular phones became very popular, reaching 30 million cell phones in 1999 and 38 million in 2003; for each 100 inhabitants, 20 have a phone, and it soon became a democratizing factor among the low income population, especially those working in the gray labor market sector. Actually, according to Bernardo Sorj (2003, 22), mobile phone is an extraordinary example of the several dimensions to be considered, when analyzing the social impact of a product or service consumed in the contemporary society. However, even today, among those who can’t have it, 50% are youngsters and the pre-paid plan accounts for a higher sales share of the companies, compared to the post-paid plan. The Map of Exclusion, a survey done by FGV (Getulio Vargas Foundation, a government funded research institute) in April, 2003, shows that information technology is still a privilege for a small portion of the Brazilian population, due to factors like financial resources, education and lack of information. From its total, 150 million people do not have access to computers, and just 26 million are digitally included. Based on these numbers, Brazil is in a worldly position very close to India and South Africa. Despite this fact, from 2000 to 2003 the access to computers grew about 50%; in other words, every four months one million people enter the digitally included group. Domestic social inclusion data depicts 12,46% of the Brazilian population as having access to a PC at home, and 8,31% to the Internet; 8,98% up to 15 years of age and 13,44 from 20–25 years of age, and 12,42% located in urban areas. Men are 48,89% of the users and in average terms, the age is 31 years old and the educational level is 8,72 years of study, relatively high if compared to the whole Brazilian population – 4,81 years. The digitally excluded group displays a different scenario: men are 49,25% of the users, the average age is 28 years old and the educational level is 4,40 years of study. The relation between digital exclusion and income becomes clear from the information that 95% of the income tax returns are filed through the Internet, since lower income individuals are not required to file. According to UNESCO, geographic disparities in literacy, low level of resources (below the necessary), which are also poorly spent, and very bad quality in terms of basic education are some of the main factors contributing to grade the country just above zero. A study from INEP (National Institute for Education and Research - a government funded institution) recently indicated that there was a performance decline of black students from 1995–2001; and this fact indicates that the unequal distribution among populations in Brazil is getting worse. The study also showed that black students entering school give up their studies much earlier than white counterparts, because education itself do not historically prize blacks. There is a diffuse racism that produces a 125 strong impact on black people’s self-esteem. Criticism on that report pointed to its significance, but also to the absence of regional variables that should be added in order to represent the large differences among the regions of the country. Research on IT and Brazilians has been conducted in recent times and digital inclusion has been discussed exhaustively. Rio de Janeiro has a young population of 412 395 inhabitants between the ages of 16–19 and, in the past few years, IT revolution has been responsible for the introduction of new products as well as new consumption habits. Mobile phones were adopted by young people for communication purposes, but also as a security device. In the 90’s, telephony system was extremely precarious, since a large part of the population did not have access to it at a reasonable cost. The Internet, on the other hand, has not been adopted as fast, due to financial barriers as well as lack of knowledge about it. Rio de Janeiro is located in the state of the same name, in the Southeastern region of Brazil. The region has one of the largest populations – more than 70 million inhabitants, a GDP per capita of R$8788.00 (4802,19 €) the highest if compared to Brazilian average of R$6486.00 (3544,26 € ), and the city is the second largest one in the region, with a population above 6 million people, of which 15,51% have access to PCs. According to a 2001 survey from IBGE (Brazilian Institute of Geography and Statistics – a government institution) São Paulo state presents the greatest inclusion in the country, with 49,7% of its schools offering access to computers while Tocantins state (in the Northern region) displays just 1,9%. In this study I kept the same structure of the questionnaire used in the original survey done in Finland in 2001 (Wilska 2002), with few adaptations to the local knowledge and reality of the observed population. A set of questions on personal security was added, due to the growing concern about urban violence/criminality in Rio de Janeiro, which contributes to a lifestyle change. According to FGV, the index of murders in Rio de Janeiro corresponds to 60 per each 100 000 people. In the 90’s IT began to extend its reach to larger segments of the society with the arrival of PCs, the Internet and mobile phones; however, in order to avoid a “digital apartheid” the State is supposed to supply, primarily, basic consumer services as water, power, and telephone fixed lines, for instance, at a reasonable cost as a way of providing for the population the needed access to new technology. 126 7.2 Theoretical Background Unlike Finland, the first ”real” information society in the world according to Castells (Wilska 2002), Brazil is a new society in terms of technology, and new configurations in terms of behavior cultures to arise; especially among young people. The industry development showed that all social strata in one way or another adopted the new media, especially the ones who work in the informal economy sector and the youngsters, because it is “cool”, eases communication, have peers’ approval, and strengthen the sense of belonging. The Internet created new ways of participating, communicating and, even, new ways of social existence, making new subjectivities arise as connected subjects in the networks; no more spectators, but actors in the virtual world (e.g. Baudrillard 2003; Morin 2003; Turkle, 1995). The mobile phone provided a sensation of freedom and privacy, among youngsters, implemented with specific symbolic meanings used in identity construction, and characterizing a ‘mobile culture’; that is, mobility is part of being young. It is to be here and there at the same time, to be connected and synchronized with the “peer group”. Alberto Melucci observes that the “action of [social] movements with youth participation is seen within the scope of symbols and communication” (Rodríguez 2002, 62). The arrival of the information technology contributed to create new consumer niches, to define new lifestyles, identities and consumption cultures, and to set out new ways of communicating. There is a “network society” generated by the information and communication revolutions, which in turn produce the expansion of PC’s, the Internet and the new broadband services for information transmission; linking phones, TV’s and computers. So, this “network society” combines production, diffusion and entertainment (op. cit., 164). Some authors have expressed negative as well as positive concerns on media technology and the changing nature of childhood; that is, of its eventual ‘disappearance’ or the technology substituting the family or the school’s role. Or even the electronic media blurring the frontier between childhood and adulthood, or robbing the true childhood (Sefton – Green 1998; Postman 2002; Wilska 2002). On the other hand, children are growing up in the computer culture and, as such, they are leading the way in this culture of simulation for the adults to follow. The “Net-Generation” as Tapscott calls it, matures earlier and shows more knowledge than the previous generations (Turkle 1995; Tapscott 1998). The fact is that computers are performing an important role in young people’s lives; however, factors like social inequalities and economic factors, are determinants on who is going to be included or excluded in the information society. 127 Theorists like Beck (1992), Giddens (1991) and Bauman (1988) have already observed that through messages of goods and practices people possess and display, they are, actually, defining themselves. Appearances are manipulated; therefore, creating and sustaining a self-identity (Warde 1994, 878). Bauman also referred to the intrinsic volatility and instability of all or almost all identities, as the ability of “going shopping” in the supermarket of identities, which is the true way for the accomplishment of fantasies regarding them [identities]. This ability would allow one to easily create and undo identities (Bauman 2000, 98), but they might also become ‘free-floating’ signifiers and group identities could fragment, turning into temporary and ephemeral signs. In this case, individuals would identify with each other through shared lifestyles, or to any kind of subculture, which Maffesoli called ‘neotribes’ for example (Gabriel – Lang 1999, 89-93). The attachment to a group or the dressing of a particular lifestyle would describe anyone instead of his/her personal characteristics, and this relates to the way consumption became associated to the self and personality. To this respect, Munro (1996) talks about a “production model of the self” which would imply in the acquisition of goods as well as services as central to one’s psychological well- being. So, constant consumption would contribute to self-growth and the like (Edgell – Hetherington – Warde, 248–273). Many questions remain, as post-modern theories want us to see consumers as just identity-seekers. According to Gabriel and Lang (1995, 93) identity should deserve and get the respect of others and lead to self-love; otherwise, it is pointless and the debate around identity fragmentation and the consumer in search of an identity might turn into meaningless clichés. In this particular survey, I tried to analyze youth lifestyles and behavior regarding consumption and its relation to technology, in order to question if new consumption ways arise among young people, and how, since they are growing in an electronically mediated culture; therefore, reinforcing or subverting the existing consumption culture. 7.3 The Survey This paper and the empirical research are based on the original survey Consumer Cultures of Young People in the Changing Information Society, carried on in Finland in 2001 (Wilska 2002) and reproduced in Rio de Janeiro, Brazil. It included the data collected between February and May 2003 with a sample of 736 students, 16–20 years of age, from private and public middle schools as well as first-year university level, and the average age was 17,2 years old. There were two main objectives: one, to compare data from both 128 countries due to its extraordinary social, economic, and cultural differences; second, to investigate attitudes towards consumption in several aspects in concern to Brazilian young people, specifically, Rio de Janeiro’s young population. There are a few Brazilian academic studies dealing with youth public policies or youth violence, but almost none exploring the issues related to consumption, particularly to technology in the contemporary society. The translation of the questionnaire followed the original, and the exception was a set of added questions on personal security and consumption, and the use of environmental logos adapted to local knowledge. As with any other western society, Brazilians, and in particular the youngsters, are witnessing an ever changing revolution in terms of information technology and mobile communication, which is responsible for new kinds of consumption, lifestyles, and consumer behavior. Marc Gobé calls the generation Y, the ‘warp-speed generation’ as they go faster and do more than the previous generations (Gobé 2001, 20). As this change in social, economic, and cultural patterns is quite new, however progressive, I tried to focus my attention on whether this new technology is already affecting the way young people express their own selves, or construct their own identities. Another argument regards the emergence of new consumer cultures, and if so, these cultures imply gender differences? An additional concern questions if the information technology will be an obstacle for the inclusion process due to factors like education and income, or it will include people socially as well as culturally along its fast development. 7.4 Work, Money and Consumption – Young People and Families Youngsters, in this survey, regarded their families’ financial situation as moderate (59,9%) and a very small percentage (1,4%) considered the family’s financial resources as insufficient. The perception of their own situation showed that 40,3% had a moderate situation and 11,4% insufficient resources. More than half of the respondents (65,7%) live with both parents and a significant number live (27,6%) just with the mother (custody is awarded to mothers in most cases). An insignificant number of students live by themselves (1,9%) which can be explained by the Brazilian family structure – children stay in parental homes for an extended number of years because it is convenient, and also due to the high unemployment rate, particularly for young people. Most of them do not work at all (69,1%) and only 3,8% work regularly; consequently, parents are responsible for children’s main expenses, as follows: house for 87.2% of the respondents, food for 91,1%, health for 87,4%, clothes for 78,7% and education for 83,3%, ranging from ‘all’ to 129 ‘most’. A recent newspaper article (April 2003) indicated that 3,6 million youngsters from 15–24 years old are looking for jobs. Parents are also in charge, financially speaking, of all leisure activities, not considered basic: hobbies for 67,9% of the respondents, social life for 72,3% and travel for 76,1%. 71,4% of parents pay for all or most of the mobile phone expenses and 14,5% pay none of them (Table 1). Mobile phone serves a dual function: parents give it to their children as a personal security device and also to control them when they go out. In general, parents choose the pre-paid plan (cards with pre-fixed amount of money), as it keeps cell phone costs under control. The three most relevant daily expenses reported were entertainment/leisure, clothing and board (26,42%, 20,59% and 17,47% respectively). Table 1. Percentage of Expenses of the Following Categories Paid for by Parents All Most Half Little Nothing Housing 82,1 5,1 4,4 3,4 5,1 Food 79,0 12,1 5,2 3,1 0,6 Health care 83,3 4,1 2,9 7,1 2,5 Clothes 59,6 19,1 7,4 12,2 1,7 Car/transport 71,5 7,8 5,8 10,2 4,7 Mobile phone bill 66,1 5,3 3,6 10,5 14,5 Studying 77,9 5,4 3,9 6,3 6,5 Hobbies 56,4 11,5 10,9 12,7 8,5 Social life, entertainment 58,7 13,6 7,9 14,6 5,2 Travelling 66,1 10,0 5,5 9,6 8,9 Rio de Janeiro is a service-oriented city, with a small number of industries, except for the pharmaceutical and oil industries. Among the most common professions, it was reported 10,7% for engineers, 7,5% for businessmen, 5,7% for military officers, and 2,6% for medical doctors. The leading occupation for mothers is being a housewife (17,7%), followed by teachers (12,6%), medical doctors (4,9%), businesswomen (3,5%), bank clerks (2,3%). Youngsters do not acknowledge other people’s influences on their consumption or purchase decisions. Parents might be the exception, with respondents showing that they either “Frequently” or “Almost Frequently” (51,9%) do influence their decisions. Beside parents, news/magazines, friends, siblings and significant others have some influence, with 29,6%, 29,4%, 25,4% and 24,2% in the first two categories. They rejected the influence of sales personnel, music videos and celebrities; on the other hand, respondents exert some influence (34,7%) on parents’ consumption; a similar rate. However, 52,2% responded “Never” or “Almost Never” for friends’ influence, 130 which is a bit surprising indicating that peers’ influence is not as strong as it is generally reported. Several studies as Childers and Rao (1992), for example, have noted that peer reference group may have a great importance in determining consumer decisions of their friends, as well as teenagers’ preferences for products and brands are influenced by peers (Gunter – Furnham 1998, 22–29). In terms of pocket money, half of the respondents received a monthly allowance, averaging R$115,20 (35,93 €) per month; however, the other half did not get an allowance. The explanation may be the fact that they ask parents for money when they need it, without a monthly amount. Leisure time expenses averaged R$66.30 (20,71 €) - trimmed mean. The majority of the respondents were able to save money (85,5%) and almost half of them (49,4%) had a particular purpose. Among the main purposes, 23,5% was for personal use; it seems that the purpose seems to be for them to use at their discretion, without parents interference. On the other hand, 14,5% did not save any money and 50,6% did not have any specific purpose. In this survey, students were asked about the existence of durable goods/electronics in their households, as well as commercial services used by them and/or their families. Technically, households are reasonably equipped with the latest products as DVD, MP3, and digital cameras (these two are not produced in the local market; therefore, they are expensive) among others. 77% of the families have a car (average 1,3 cars), and 95% have at least one washer in the household. TV sets are also overwhelmingly present, with a percentage of 100% of all households, averaging 3,3 TVs per household. Home computers are being added to the existing durable goods at a fast pace, and in this study 89% have at least one, and 81% have one Internet connection at the minimum (Table 2). Youngsters’ own spaces are fairly equipped with electronics; both CD player categories (table/portable) had a relevant percentage – 91% and 62%. Only four youngsters had their own car, all others acknowledged the family car as theirs. 131 Table 2. Ownership of Durable Goods in Youngsters’ Households Households (parental) owning at least one (N=736) (%) Average per household Youth owning at least one (N=736) (number) Youth owning at least one (N=736) (%) Car 77 1,30 4 0,5 Boat 4 0,05 2 0,3 Motor bike 11 0,13 29 3,9 TV 100 3,34 170 23,1 VCR 94 1,53 84 11,4 Cable TV 63 1,14 61 8,3 PC 89 1,25 191 26,0 Laptop 12 0,14 6 0,8 Games 65 0,96 235 31,9 Internet connection 81 1,03 144 19,6 Wireless Internet 12 0,18 44 6,0 Palm Pilot 6 0,07 8 1,1 Video camera 40 0,45 19 2,6 Digital camera 18 0,21 27 3,7 CD Player 91 1,81 229 31,1 Portable CD 62 0,83 237 32,2 DVD player 43 0,51 38 5,2 MP3 17 0,20 44 6,0 Washer 95 1,05 5 0,7 Dishwasher 26 0,28 0 0,0 Microwave 71 0,77 4 0,5 Coffee maker 69 0,81 2 0,3 Regarding commercial services (Table 3), one has the impression that some respondents would choose an option in a random way, for being ashamed of not using the service, or maybe upset for being asked. This conclusion is based, sometimes, on the family’s income or the father’s occupation that did not allow certain expenses. The most common services listed under the word frequently were related either with entertainment/leisure, sports practice/gym, health insurance or education/language instruction; the ones used sometimes gives a different picture, with higher percentages for almost all categories with few exceptions. It means that families’ income allow several expenses from time to time, not at a frequent basis. The category never shows that more expensive services have high percentages as spa (85%), travel abroad (59%), concerts (58%) or even car wash (25%). 132 Table 3. Commercial Services Use – Young People/Family (%) Frequently Sometimes Never Packaged tours abroad 3 30 59 Tours in Brazil 7 46 42 The use of taxi 29 60 9 Movies 62 35 2 Eating out 43 53 3 Cafes, fast-food restaurants 48 44 5 Bars, pubs 42 52 5 Discos, night clubs 22 50 24 Theatre 10 55 30 Concerts 5 27 58 Exhibitions 8 56 30 Fairs 12 61 22 Sports services 39 44 13 Private education courses 53 27 16 Private leisure time courses 31 37 26 Private health care 59 19 17 Gym, massage,etc. 38 29 28 Beauty treatment 18 43 32 Spas 1 5 85 Private child care 4 7 80 House cleaning 24 18 50 Sewing, dress making 4 36 52 Catering services 3 31 58 Renovating services 8 67 19 Machinery repair and maintenance 13 71 11 House move services 3 42 46 Interior decoration services 7 31 54 Garden planning 3 14 73 Gardening, planting 5 17 71 Car repair and maintenance 19 58 19 Car wash 32 40 25 133 7.4.1 Respondent as Consumer/Individual Consumption Respondents were asked to analyze themselves as consumers in some categories as individualism, impulsiveness, trendiness, thriftiness and environment consciousness and their opposites (Table 4). The percentages show moderate results with no extremes at all. Using a scale of 1 to 5 (1= agree with statement at left, 5= agree with statement at right, and 2, 3, 4= partially agree) the numbers ranged from 1,99 to 4,11, that is, students placed themselves in the middle categories. It means that respondents were able to save money, but they paid attention to new trends. They did not show themselves as impulsive buyers and considered themselves neither as mass consumers nor as individualists and, they indicated some kind of concern for the environment. Table 4. Percentages of self-perceived consumer typologies – 1–5 scale Mean 1 2 3 4 5 Money slips through my fingers 4,11 5 5 16 22 52 I manage to save something Trend-conscious 2,91 13 22 40 13 12 "Laggard" Impulsive buyer 3,74 5 10 24 26 34 Prudent Consumer Mass consumer 3,27 6 13 47 17 17 Individualist Environmentally conscious 1,99 42 27 24 5 3 "Free Rider" It is worth noticing that there were gender differences when youngsters stated their preferences in this question. In the trend-consciousness dimension, twice as much girls than boys declared to be trend conscious, and the opposite happened in the other side of the scale. This shows that girls are more willing to admit that they are trend conscious than boys. In the dimensions impulsiveness and thriftiness girls displayed a higher percentage in the extreme categories. 7.4.2 Mobile Phone and Consumption Styles Since mobile phone became an essential part in people’s lives, the same can be said in respect to youngsters. It was never a simple gadget, since it became a feature of one’s lifestyle as well as one’s identity, since it was easy to 134 personalize with logos, tones, and accessories. It is now an intrinsic part to young people’s consumption, not necessarily a different kind of consumption. According to the results, 78,7% had a mobile phone; for 93,5% of them it was either the first, second or third phone (trimmed mean 1,9 mobile phones). Two years and a half was the average time of use, the same as in Finland (see Wilska 2003), and the bills averaged R$38,50 (2,03€). It is important to notice that 63,7% prefer the pre-paid plan (card) and the other 23,4% of the respondents shared the post-paid (bill by the end of the month usually paid by parents), limit of calls 9% and other plan(s) 4%. Respondents reported that they receive an average of 3,66 calls and also make an average of 3,11 calls. In respect to text messages they receive 1,61 and send 1,27. Both results are far inferior to the ones found in the Finnish study (see Wilska 2002). The most relevant phone services used by the respondents include: call wait (33,4%), home line (61%), mailbox (66,7%), text messages (86,6%), alarm/calculator (81,4%), new logos (68%) and text message service (80,9%). A Likert scale 1–5 was used to measure attitudes (1=strongly agree, 5=strongly disagree). Attitudes were then analyzed by factor analysis, using principal component extraction (table 5). Factor analysis does not test hypotheses using a large set of variables, it tries to summarize the data using a small set of components. In this set of questions describing the mobile use, a three-factor solution was also used in order to ease the data comparison. The principal components, Addict User, Fashion Oriented, and Practical/Thrifty Oriented explained about 38,65% of the total variance. For the first factor, the Addict User, attitudes emphasizing intensive use of cell phone as a gadget got high loadings. It means that this young individual writes many messages, and likes to get phone calls as well as text messages; sometimes makes phone calls with no particular purpose. He/she uses the mobile in public places and feels uncomfortable if the phone is not with him/her, and in the future, he/she will use even more the mobile communication. For a person like this, it might be difficult to manage the phone bills, since more use is always preferable to less. The second factor, the Fashion-Oriented Individual, looks for trendy models on the cutting-edge of the latest technology. At the same time, it has to fit a particular lifestyle, the individual’s image. It is important to receive many calls and text messages and to display an Internet connection. This young person also feels uncomfortable if the mobile is not around and he/she cares about the operator’s image – which one is more fashionable? On the other hand, he/she is not worried about the environment, as far as he/she is not concerned about what happens to old mobile phones. The third and last factor shows the features of a practical and thrifty young individual. For him/her a mobile phone is not a gadget; it is to be used when one needs it, and does not 135 have to be fancy as price factor is something to be considered. This device will be used even more in the future, because of its countless advantages. 136 Table 5. Factor Analysis of Mobile Phone Use and Young People Factor 1 Addict User Factor 2 Fashion oriented Factor 3 Practical / Thrifty h2 Mobile Phone is necessary only for connecting people and organizing things 0.514 0.311 I often make phone calls without any particular purpose 0.448 -0.385 0.377 I write many text messages 0.624 0.438 I use very much the fixed line at home 0.373 0.155 The cheapest model is good enough for me -0.607 0.328 0.492 Three years old mobile phone looks too old-fashioned for me 0.734 0.546 It is important for me that the mobile phone is the latest technology and "posh" 0.808 0.667 I often change logos and/or ringing tones 0.464 0.274 It is important for me to receive many phone calls and text messages 0.631 0.404 0.565 I often use my mobile phone in public places (bus, street, cafes) 0.583 0.346 It is important for me an internet connection on my mobile phone 0.509 0.314 It is important for me that my phone fits my clothing style and general image 0.620 0.463 I keep checking for possible phone calls and text messages all the time 0.690 0.556 I feel very uncomfortable if, for any reason, I don't have my phone with me 0.583 0.359 0.470 I often have difficulties in paying my mobile phone bills 0.313 0.121 I don't care with what happens to old mobile phones 0.363 0.164 I will probably use the mobile phone even more in the future 0.423 0.367 0.327 Price is the most important issue when choosing a phone 0.703 0.510 Some operators or connection types are more "trendy" than others 0.429 0.247 Eigenvalue 3.032 3.011 1.300 Explained (%) 15.956 15.846 6.844 38.646 Kaiser-Meyer-Olkin Measure of Sampling Adequacy.= 0.855 Bartlett's Test of Sphericity = 1946.22 Sig=.000 Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. a.Rotation converged in 5 iterations. These three “dimensions” can be used to analyze individuals. For example, one particular person can be highly “Fashion Oriented”, meaning that he/she is deeply aware of appearance, but also highly addict to the phone as a gadget. It 137 is a way of finding non-observed underlying dimensions of human behaviour and using this set of questions to measure them. The next logical step is to further analyze the data, by creating and saving standardized sum variables representing the three dimensions, which were weighted by the loadings of at least +/- .30, and checking how independent variables as gender and individual attitudes towards consumption affected them. Separate ANOVA analyses were run for each factor, but a low explanatory power of the independent variables was found (maximum of 10,2% for the first factor). Gender, as well as a score representing individual consumption attitudes (Trendiness, 1-Trend Conscious, 5- Laggard), a self assessment of the individual’s wealth (1-Very much, 5-Very sparsely) and the type of school (Private, public) were used as independent variables attempting to explain the factors. Gender was statistically significant for the Addict Factor, at the p<.001 level, however, with a small to moderate effect of 2,4% (eta squared). Interestingly, it showed a negative value for boys, relative to girls, meaning that boys seem to be less addicted than girls. Back to the data, girls tend to “feel uncomfortable when they don’t have their cell phones on them”, they write more text messages, they “keep checking for messages and phone calls” and “often change logos and tones” to a higher degree than boys. Trendiness was also significant for this factor, at the p<.05 level, with a moderate effect (5,1% eta squared), with trend conscious subjects being more “addict” than “laggards”. The independent variable found most statistically significant for the Fashion Oriented factor was Trendiness, at the p<.001 level and a moderate effect (6.1%), with the most self-declared trend conscious subjects being more Fashion Oriented. Self-declared perception of wealth (“do you have money?”) also plays a statistically significant role, with students that have “very much” money being more Fashion Oriented. The third factor, described as Practical/Thrifty had only 6,2% of its variance explained, and presented a significant relationship for the self declared perception of wealth (at the level of p<.01), with the higher end of income being less practical/thrifty than the lower end. Another significant independent variable for this factor was the type of school, with students from private schools being less practical/thrifty than the ones from public schools. Both explanatory variables probably work as proxies for wealth, meaning that youngsters from higher income families pay less attention to practicalities and to the low price of the products/services. Comparing the results of the Finnish and the Brazilian surveys, it is interesting to note that gender was not significant for the “trendy” dimension, apparently indicating that boys and girls are equally inclined to observe trends and fashions in Brazil; while in Finland (see Wilska 2002), girls displayed less of that tendency than boys. Another important difference was the variable 138 “impulsiveness” affecting with high statistical significance all the three dimensions in Finland, while being not significant in Brazil. Environmental consciousness was also not important in Brazil, but it was at the p<0.05 level for “trendy” and “thrifty” dimensions in Finland. Table 6. Connection Between Mobile Phone Use Styles, Gender and Consumer Types Analyzed by Multiple ANOVAs Factor 1 Factor 2 Factor 3 Addict User Fashion oriented Practical B B B 1.Gender 1 Boy 159 -0.31 2 Girl 168 0(a) F 7.66** 2.95 1.31 54.Trendiness 1 Trend conscious 50 0.53 0.57 2 74 0.23 0.19 3 125 -0.07 -0.01 4 43 -0.15 -0.30 5 Laggard 35 0(a) 0(a) F 4.28** 5.10*** 0.60 24.Do you have money? 1 Very much 8 1.08 -0.83 2 Much 65 0.50 -0.43 3 Moderately 144 0.24 -0.27 4 Sparsely 84 0.45 -0.71 5 Very sparsely 26 0(a) 0(a) F 1.58 2.90* 3.62 Type of school 1 Private 224 -0.29 2 Public 103 0(a) F 0.11 2.92 5.18 100R2 10.2 10.0 6.2 a. The value is zero, because it is a reference category p <0.001*** p <0.01** p <0.05* 7.5 Young People, Information Technology and Consumption The new technology brought computers and along with them, the Internet, which transformed the way we are, the way we think and behave, and created an entirely new way of communicating. Edgard Morin (2003) talks about interconnected individuals in the networks, where exchanges, reciprocities and 139 liberties are bound to happen. In post-modern societies, identities are no more fixed structures; they move freely or, as Featherstone puts it, people do not look for a single identity or image but as many as possible to match different moods (Featherstone 1991, 27). And the Internet became an important social lab for experimenting with the constructions and reconstructions of the self according to Turkle (1995, 180). For the young generation the Internet as well as the mobile became essential parts of their lives. The results showed that 40,5% used the Net for some time per month, 21,3% once a day, and 0,6% never used. Out of 736 respondents, 104 reported they do not have a PC; 87,8% had access at home and the other small percentages are distributed among accessing at school (4,3), friend’s house (3,1), and somewhere else (4,0). The preferred sites included search for information (99,2%), entertainment (95,3%), email (93,7%), chat/ICQ (83,2%), and newspapers/magazines (72,5%), all ranging from ‘frequently’ to ‘sometimes’. In certain way, one finds out that young people master quite easily the transformations that occur in society and, especially with the digital revolution, one concludes that they are accomplished and sophisticated users. The Internet as a social space also attracted the youngsters; 54,2% mentioned that they made friends through the Net, and 45,8% not at all. The same technique (Factor Analysis) was used to measure the relation between mobile phones and lifestyles was applied in this section - Youngsters and Information Technology. Attitudes were evaluated using a Likert scale (1=strongly agree, 5= strongly disagree) and after that a factor analysis statistics was performed (Table 7). With a Varimax Extraction, the solution chosen was a four-factor analysis, and the result explained 46,8% of the total variance. The factors were Skilled User, Suspicious, Pessimistic, and Gadget Enthusiast. The first factor showed high loadings for the statement “I am very skilled user of IT” and for other questions expressing enthusiasm for, willingness to work with and praise to the importance of information technology. The suspicious user finds IT uninteresting, is against it, and neither understands nor wants to learn about it, being afraid that he/she will drop out of its fast development. The pessimistic sees environmental problems, undue interference, increased inequality and high cost associated with IT. Finally, the gadget enthusiast sees IT as a solution for environmental problems, likes the technology, while not being an expert on it, and thinks that it is expensive because he/she would like to use more of it. 140 Table 7. Attitudes towards IT and information society Component 1 2 3 4 Skilled User Suspicious Pessimistic Gadget Enthusiast h 2 IT is unavoidable, but is not interesting 0.417 -0.336 0.321 I am a very skilled IT user 0.399 -0.431 0.437 I am against IT 0.546 -0.374 0.444 I would like to get a job with IT 0.638 0.449 I am very enthusiastic about tech innovation 0.437 -0.384 0.393 0.493 I use IT very much in social life 0.376 -0.386 0.326 IT rules our lives too much 0.623 0.398 IT makes our life easier 0.465 -0.312 0.421 IT increases environmental problems 0.703 0.501 IT increases inequality 0.679 0.470 I don't understand anything about IT 0.746 0.591 I am sick and tired about all the fuss on IT 0.753 0.614 IT may produce solutions for environmental problems 0.534 0.331 I am afraid I'll drop out of rapid development of IT 0.689 0.528 It is important for me to own the latest technical innovations 0.707 0.526 I will get a digital TV as soon as possible 0.650 0.516 I will use IT a lot in the future 0.628 0.477 IT and digital devices are too expensive 0.408 0.580 0.503 I dream of getting rich in IT business 0.728 0.549 Eigenvalue 3.064 2.744 1.730 1.355 Explained (%) 16.126 14.444 9.107 7.130 46.806 Kaiser-Meyer-Olkin Measure of Sampling Adequacy=0.826 Bartlett's Test of Sphericity= 2701.6 Sig=.000 Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. a Rotation converged in 6 iterations. 141 The attitude factors were further analyzed by turning the factors into standardized sum variables that were weighed by the loadings of at least +/- .30. Separate ANOVA analysis were performed to assess the explanatory importance and significance of factors like Gender, Type of School (private vs. public), individual consumption features of the respondents and a two category Consumer Index, based on the declared ownership of durable goods. Gender was significant at the level p<.001, with an F value of 54,97, for the “Skilled User” factor, meaning that boys are more “Skilled” than girls regarding IT. Type of school (public, private) was also significant, at the p<.05 level (Students from private schools are more skilled than the ones from public schools). Trendiness was a significant factor for the “Suspicious” dimension, at the p<.05 level, and the Consumer Index was significant at the p<.001 level, providing an indication that ownership of durable goods, acting as proxy of income, had a significant effect on suspicion: the lower the income, the more IT suspicious. Only Gender was significant at the p<.01 level for the Pessimistic, with boys being less pessimistic than girls, and Impulsiveness was significant at the .05 level for the Gadget enthusiast, with Impulsive subjects being less Gadget enthusiasts. In the Finnish study, Gender was statistically significant at the p<.001 level for all factors, while here, it was significant at that level just for the Skilled User, and at the p<.01 level, for the Pessimistic User. This is surprising, considering that the Finnish society has a higher degree of gender equality than Brazil, a country where the “macho” domination is normally taken for granted. It might represent, though, that other variables not considered in the study may be very relevant, somehow reducing the explanatory power of the gender. For example, the p<.001 significance of the Consumer Index variable in the Suspicious dimension might indicate the need of more variables measuring income and social stratification for Brazil. 142 Table 8. Connection Between Attitudes Towards IT and Information Society, Gender, School, Money and Consumer Types Analyzed by Multiple ANOVAs Factor 1 Factor 2 Factor 3 Factor 4 Skilled User Suspicious Pessimistic Gadget Enthusiast B B B B Gender 1Boy 324 0.56 -0.23 2Girl 330 0(a) 0(a) F 54.97*** 1.65 8.54 ** 0.73 Type of school 1Private 420 0.16 2Public 234 0(a) F 4.07* 0.01 1.26 2.44 Impulsiveness 1 Impulsive buyer 35 -0.37 2 67 -0.35 3 154 -0.25 4 174 -0.21 5 Prudent buyer 224 0(a) F 2.35 0.87 0.11 2.66* Trendiness 1 Trend conscious 84 -0.56 2 140 -0.27 3 268 -0.32 4 81 -0.24 5Laggard 78 0(a) F 2.19 3.14* 0.91 0.63 Consumer Index 1Lower 312 0.29 2Higher 342 0(a) F 2.66 13.35*** 2.90 3.24 100R2 11.9 5.1 2.9 3.1 7.6 Consumption, Security and Environmental Consciousness In this study, results have shown that security is an important issue in Brazilian large cities and, in recent years, people started to become more aware and to change ways of living due to the growing violence and fear. Just five questions gave the approximate dimension of the problem regarding, especially, young people. A cross-tabulation between genres, schools, and 143 areas of living gave the following significant results: more girls than boys changed consumption habits, like going out to any place, or anytime; more students from private schools than public ones and living in the north-center area of the city. When asked if they preferred to go out alone, it was found a higher percentage for boys than for girls, for students from public than private schools, and living in the north-center region. In relation to going out with peers, the tabulation showed the following: more girls than boys prefer the company of a group, they belong to private schools and the highest percentage comes out from the south-west area. The next question addresses the use of mobile phone to get in touch with parents and a higher percentage corresponds to girls than to boys; they belong to private schools, and they live in the southwest are (the highest purchasing power). It is worth noting that 52,2% of the youngsters from the rural area communicate with parents; however, 26,9% do not and the explanation is linked to the fact that they do not have as many mobiles as the ones from the urban areas. Respondents reported that they preferred going to shopping malls as a measure of precaution than to any other place, but results were not very different. Changes due to violence happened in all schools in a high degree; overall about half of the students changed in their habits. Changes were higher for girls, studying in private schools, and urban areas. Anyway, one observes that there was a change in terms of keeping the same habits or even lifestyles. A question on environmental consciousness was added to the survey as follows: “Consuming in a conscious way is part of your life values?” and the positive result (88,1%) was a bit surprising, because it is reported that students do not read enough, and have a limited vocabulary, and have trouble interpreting meanings, in general. Respondents had difficulty in many questions related to this topic. The majority of youngsters (53,6%) ignore if their families buy biologically grown products and they never heard about it (77,3%). It is worth mentioning that these products are arranged in a separate section at the supermarkets, but they are expensive. The ‘green products’ showed a different percentage: 20,8% do not know if their families buy it and do not know the products (64,3%); however, 46,0% buy occasionally, and 25,8% do it regularly. There is a program for recycling materials, at municipal level, but it has yet to be implemented in the whole city. The percentage ranged from 4,8% - people separate everything to 53% - do not separate anything. Students (52,4%) also reported that they do not buy pirate or second- hand clothes, and the reason for 51,2% was the suspicion about quality; however, 41,5% buy it occasionally. Other products, like CDs or books attract 52,1%. When asked about the future, the respondents depicted a very interesting picture. Around 85% of the youngsters stated that they have high hopes of 144 being well established in life, with a very good job and financial success, and 92% will likely own a house and a car. Money and material goods, the core incentive of consumer society, are on the top of the mind for these youngsters. However, family life and values, as well as tradition are still highly regarded as far as they become an important project to pursue for 72% of the youngsters. 7.7 Conclusions This survey was a reproduction of the study conducted in Finland in 2001 (see Wilska 2002), aiming at a cross-cultural comparison, and investigated the relation between young people’s consumption habits and information technology; the use of mobile phone, and the Internet; the student as an individual consumer and his/her relation to money, durable goods, environmental issues and personal security. Youngsters are incorporating, in a fast pace, both information and technology to their lifestyles; however, one cannot say that identity construction is totally dependable on IT. Characteristically of this group, youngsters are easily adaptable to postmodern and globalized times: they show free-floating identities, with interchangeable selves, making it possible to affiliate with different groups, with no commitment whatsoever and any new device is welcome. Although consumption is an important and visible component, responsible for a major change in social, cultural and economic terms in Brazilian society, few people can really have access to modern technology and material goods in general. In the first place, there is a lack of basic goods as healthcare and urban infrastructure, inequalities in income distribution and growing unemployment, which generate difficulties for the democratization of services. The factor analysis for the data of the present study was surprisingly similar to the Finnish data (Wilska 2002), with approximately the same factors found in both studies for cell phone and IT analysis, and comparable loadings for the questions. The similarity is particularly striking if we consider that the subjects of the surveys come from very different social, ethnic, economic and cultural backgrounds. The similarity may come from their age, and to the fact that they live in a global society and are exposed to the same information that comes from the Internet, MTV and consumption products. Some differences are displayed in the relationship between the factors extracted and the explanatory variables used. Some questions may have not been completely understood by the group tested in Rio, like the attitudes towards the environment, and the consumption of some kind of goods that are 145 not mass consumed in Brazil. The “welfare society” might be an explanatory factor for the differences found, and must be closely related to social stratification, economic and cultural inequalities and resulting information exclusion. A direct comparison of the survey data might be used in a possible study of the social dimensions of the information revolution, in order to explore the differences and similarities in a more thorough way. 146 References Baudrillard, Jean (2003) Seminar on Subjectivity and Digital Culture at Universidade Candido Mendes, Rio de Janeiro, June 20/2003. Brazilian Institute of Geography and Statistics. Featherstone, Mike (1991) Consumer Culture and Postmodernism. Sage: London. Gabriel, Yannis – Lang, Tim (1995) The Unmanageable Consumer: Contemporary Consumption and its Fragmentations. Sage: London. Gobé, Marc (2001) Emotional Branding: The New Paradigm for Connecting Brands to People. Allworth Press: New York. Gunter, Barrie – Furnham, Adrian (1998) Children as Consumers: A Psychological Analysis of the Young People’s Market. Routledge: London. Morin, Edgard (2003) Seminar on Subjectivity and Digital Culture at Universidade Candido Mendes, Rio de Janeiro, June/2003. Munro, Rolland (1996) The consumption view of self: extension, exchange and identity. In: Consumption Matters, ed. by S. Edgell, K. Hetherington and A. Warde. Blackwell: London. National Institute for Education and Research, Postman, Neil (2002) O Desaparecimento da Infância. (The Disappearance of Childhood.) RJ: Graphia Editorial. Rodriguez, Félix (ed.) (2002.) Comunicación y Cultura Juvenil. Editorial Ariel: Barcelona. Sefton-Green, Julian (ed.) (1998) Digital Diversions: Youth Culture in the Age of Multimedia. UCL Press: London. Sorj, Bernardo (2000) A Nova Sociedade Brasileira. (The New Brazilian Society.) Zahar: Rio. Tapscott, Don (1998) Growing up digital: the rise of the net generation. McGraw Hill: New York. Turkle, Sherry (1995) Life on the Screen: Identity in the Age of the Internet. Simon & Schuster: New York. Warde, Alan (1994) Consumption, identity-formation and uncertainty. Sociology, Vol. 28, No. 4, 877–898. Wilska, Terhi-Anna (2003) Mobile Phone Use as Part of Young People’s Consumption Styles. Journal of Consumer Policy. Forthcoming. 147 Wilska, Terhi-Anna (2002) New Technology and Consumption Styles among Finnish Young People. Paper presented at the Nordic Sociological Conference, Reykjavik. Internet-sites: www.folha.com.br 148 8 THE CHANGING CHARACTERISTICS OF COMPUTER USE IN FINLAND 1987-200015 Pekka Räsänen Abstract The paper addresses the changing characteristics of computer use in Finland between the late 1980s and late 1990s. The starting point is in the conflicting symbolic value of information technology (IT). It is assumed that computing represents a problematic arena for culturally distinctive consumption because there is no clear value of ‘high’ culture attached to it. Despite this, however, those who adopt new innovations faster than the general consumer market can be seen as pioneers who are able to make sense of the relative advantages of new technology. But what are the structural differences of computer use and how they have changed over the last decade? Empirical analysis is based on data from the time use surveys conducted by Statistics Finland in 1987/88 and 1999/2000. On one hand, we are able to observe a relatively clear socio- demographic pattern of computer use. On the other hand, the observed pattern cannot be easily interpreted in the light of basic social and cultural hierachies. KEYWORDS: Time use research, computer use, information technology, socio-demographic structures. 8.1 Introduction The Nordic countries share many characteristics that have been associated with ‘information societies’. Particularly the role of new communication and other information technologies is often seen as the implicit determinant of the information revolution (see Castells – Himanen 2002; Viherä – Nurmela 15 This paper is a slightly modified version of the paper presented at the 25th International Association for Time Use Research (IATUR) Conference in Brussels, Belgium, 17-19 September 2003. 149 2001). In the Nordic countries traditional manufacturing has developed rapidly toward technology-driven industries and innovation-based economies. The diffusion of the new information technologies can be recognised also at the individual level. International comparisons show that people are especially well equipped with mobile phones, computers and Internet connections. Measured by the penetration of these appliances, the Nordic Countries are, in fact, the leading ones in the OECD (Nordic Information Society Statistics 2002). This paper discusses the use of information technology (IT) in Finland. The paper concentrates on the diffusion of home computers into the Finnish households and compares time use on computing between the late 1980s and 1990s. It is examined what kind of socio-demographic differences there are in the possession of home computers and the patterns of time use on computing. Can we find any differences by economic activity, socio-economic status, age, gender and educational level? It is also asked how these structural differences have changed between the late 1980s and 1990s. Data from the Finnish Time Use Surveys of the year 1987/1988 and 1999/2000 are used in the empirical analyses. Both of these data sets are based on time diary studies and they are collected by Statistics Finland. In time use surveys respondents are asked to fill in a diary for two days (one weekday, one weekend day). They are asked to record, in their own words, their primary activity, and what else they were doing at the same time.16 In addition to time-use questions, data sets include information about respondents’ socio-demographic characteristics, accommodation and about the possession of various durables. Some of this information is also utilised in the paper. The utilised time use surveys cover Finnish population aged 10 years and over. The data aim to describe all central time use categories that do not vary much by day of the week or population group. The number of cases (days investigated) is approximately 15 400 in 1987/1988 and 10 500 in 1999/2000. (Niemi – Pääkkönen 1990, 11–12; Niemi – Pääkkönen 2002, 111–112.) 16 The record keeping was done on a 10 minute-basis. If respondents are doing several things at once, they are asked to specify their ‘main activity’ and ‘secondary activities’. In this paper, the attention is confined to ‘main activity’ alone. 1987/1988 data was collected between April 1987 and March 1988; 1999/2000 data between March 1999 and March 2002. 150 8.2 Approaching computer use from a cultural perspective There is a conflicting symbolic value related to IT in the present day societies. On one hand, many ideas about the privileged role of information specialists in the work and stratification research have been presented (see Bell 1973; Berkeley 1971; Drucker 1968). It is assumed, for example, that the ‘technocratic’ power of information specialists can challenge the old managerial-based authority in the work-places. Their professional authority is considered to be based on the task, not on the managerial rank (Lewin – Orleans 2000). But at the same time, leisure-time computing as such does not represent a typical pattern of culturally distinctive consumption. This is because there is no clear value of ‘high’ culture attached to it. The use of new technologies or other computer-related activities cannot easily be associated with the activities of the upper classes similarly as, say, opera, theatre or literature. Despite this, those who adopt new innovations faster than the general consumer market can be seen as pioneers who are able to make sense of the advantages of the new technology. Many factors can affect how uses for innovations are perceived. Everett Rogers (1986), for example, has proposed that the characteristics of new products, as perceived by members of society, determine the rate of their adaptation. According to him, there are a total of five important attributes of an innovation. These are: 1) relative advantage, 2) compatibility, 3) complexity, 4) reliability, and 5) observability. In Roger’s view, these five attributes explain the success of any new form of technological innovation; how comfortably and how easily it fits into people’s life. Other writers have stressed that also familiarity is an important attribute of innovations considering the easy adaptation of new products or services within consumer market (e.g. Dickson 2000; Fidler 1997; Räsänen 2003). In general, familiarity refers to the bridges that the earlier innovations have provided. New forms of doing something are not often adopted without familiar links to earlier or existing forms of doing something rather similar. Successful innovations do not emerge from nowhere. This means that each new innovation is assumed to “evolve from its origins as a recognisable extension of an earlier form into distinct form all its own” (Fidler 1997, 16). Besides the mentioned important attributes, also other factors influence the adaptation of new products. Economic and socio-cultural resources of individuals can be understood to be among the most important factors. It can be argued that when certain consumer goods are distributed in the general market, there are structural conditions that determine the rate of their adoption among different population groups. It is often possible to identify different categories of consumers, such as early adopters, the majority, and late adopters 151 by using typical socio-demographic characteristics. The penetration of mobile phones, for example, was highly influenced by the socio-economic status of household in Finland. Mobile phones were first adopted among the entrepreneurs, professionals and managers (e.g. Nurmela 1997; Räsänen 2003). Similar differences have been found also in the study of computer and Internet use. The access rate of Internet and computers has risen significantly faster among clerical employees and entrepreneurs over the 1990s than among other socio-economic groups (see Nurmela et. al. 2000). New innovations and products are in many cases adapted by the members of society in particular phases. First, there are usually strong status distinctions when an innovation is introduced. Later, as the new products or services become more commonplace, the links between social or cultural resources and consumption of these products or services begin to diminish. However, it has also been noted in sociological literature that the proliferation of consumer goods does not necessary weaken the connections between consumption behaviour and social background (e.g. Bourdieu 1984; Kraaykamp 2002). Certain patterns of consumption may remain distinctive for certain groups of people for relatively long periods of time. Such culturally determined differences, which seem to reproduce over time, may be related to many activities. Thus an examination of the differences between socio-economic positions and other socio-demographic factors can be seen as relevant. Next, the basic socio-demographic features related to the possession of home computers in Finland will be discussed. What kind of differences we can find when the early penetration of home computers in the 1980s is examined by labour market position, socio-economic status, age, gender and educational level? Can we observe similar socio-demographic differences also in the late 1990s, or have home computers become evenly diffused into Finnish households? 8.3 Possession of home computers in the late 1980s and 1990s In 2002, approximately 75 percent of Finns were able to use a computer, whether at home, work, school or somewhere else. At the same time, about 62 percent had a computer at home. (Nordic Information Society Statistics 2002, 13-14.) While the penetration of home computers is now rather high in the Finland, this was not situation in the 1990s. In the mid 1990s, for example, about 30 percent of the Finnish households possessed a computer (e.g. Nurmela 1997, 15; Raijas 2001). In the late 1980s, only 17 percent of respondents had a computer in their household. Ten years later, however, 55 percent of the respondents had a 152 computer in their household. These proportions are based on the Finnish Time Use Surveys 1987/1988 and 1999/2000.17 The figures suggest that there have also been considerable socio-demographic differences in the possession of computers both in the late 1980s and 1990s. We can assume this simply because computers have penetrated only a part of the Finnish households. But which factors explain the possession of home computers? Diffusion of home computers by the basic socio-demographic characteristics is presented in Table 1. As table shows, there were considerable differences both in the 1980s and 1990s by all the selected characteristics. Let us first take a closer look at the differences by labour market position and socio-economic status. Households labelled as ‘other’ in the classification represent mainly those on pension and sickness allowance. As table shows, computers have become more frequent in all labour market and socio- economic positions. In both years the possession of a home computer was most common in student households; it is most uncommon with those who perform domestic work. Examination by socio-economic status reveals that computer access has been most common among other entrepreneurs than farmers and among upper level employees (professionals and managers). In 1987/1988, less than six percent of the farmers had a computer. Ten years later, however, more than half of the farmers had a computer at home. This result is, of course, affected by the changes in the structure of the employed labour force in ten years, but it also shows that computers have become rather widespread in all socio- economic groups. Otherwise, no interesting changes between the years can be observed in these groups. Possession of home computer was most rare in manual worker households in both years, with the exception of farmers in 1987/1988. At that time, only five and a half percent of farmer household had a computer. Comparisons by age, gender and level of education show similar results. In 1987/1988, a computer was most common in the youngest age group. This has not changed in the 1990s. Women had fewer accesses to computers than men in both years, even though the difference has balanced considerably since the 1980s. Differences by level of education indicate that the most educated people possessed more computers than others in the late 1980s. This pattern has become less obvious but education appears to matter also in the 1990s. Half of the respondents with only lower secondary education had access to 17 The proportions indicate how many percent of the respondents represented in the data have a computer access in their household. Thus, the percent values do not offer us the accurate estimations of total Finnish households possessing a computer at the given points of time. 153 home computer in 1999/2000. At the same time, more than 80 percent of those having Master’s degree had a computer at home. Table 1 Possession of home computers in Finland by labour market position, socio- economic status, age, gender and level of education in the late 1980s and 1990s (%) 1987-1988 1999-2000 N Labour market position Employed 16,0 61,6 1476/3381 Unemployed 8,4 40,6 28/228 Homemaker 9,6 44,6 38/139 Other 2,8 16,9 64/335 Student 33,0 77,4 1025/1711 Socio-economic status Farmer 5,5 53,9 41/172 Other entrepreneur 24,5 65,2 171/238 Professional, manager 24,9 81,3 366/946 Intermediate non-manual 16,1 62,6 514/1140 Manual worker 13,4 47,4 464/778 Age 10-17 years 37,6 79,8 905/1149 18-30 years 11,3 59,2 418/1221 31-45 years 21,5 66,5 952/1690 46-50 years 10,6 54,3 324/1464 Over 60 years 1,8 14,8 32/269 Gender Male 20,6 57,1 1527/2806 Female 13,9 52,9 1104/2988 Level of education Lower secondary 14,1 53,6 840/1902 Lowest upper level 23,2 64,4 127/895 Bachelor 22,7 76,6 70/416 Master 25,7 82,9 135/435 Ph.D. 31,3 83,3 15/50 Total 17,1 54,9 2631/4797 The examination of the possession of computers indicates that Finnish people have indeed become well equipped with home computers. Differences in the computer access are not very large; but there still exists differences 154 between various household characteristics. The clearest differences can be observed by labour market positions, by socio-economic status and by age. However, these differences do not tell us anything about the differences of actual computer use between social groups. Next, the use of computers is examined by comparing the average times that people spent on computing.18 What kind of differences we can find by examining computer use in this way? 8.4 Time use on computing in the late 1980s and 1990s As a time-use category, computing can be considered as a typical free time category. Free time means generally residual time: that time which is not spent on earning money or on housework or personal care. It thus consists of all kinds of activities that we engage in outside working hours. For example, one can play games, listen to music, or even compose music with computers. Computers can also be used for communication purposes if there is an Internet access available. The World Wide Web is perhaps the most popular computer- based commercial service in the advanced societies (e.g. Dickson 2000; Franzen 2000). From the latest sport news and gossip columns to chat boards; almost anything that can be presented digitally has become accessible through net. But in addition to various leisure-time purposes, computers are commonly used for gainful employment. Most remote workers all over the industrialized societies are working via computers (e.g. Florida 2002; Blom et. al. 2002). In many cases it is enough to have another clerical employee (or a research colleague) available over the phone or Email. In a way, computers are among the main reasons why remote work has become a realistic option for many companies. Besides this, it is true that many people work in the evenings and at home; these people are typical computer users. Of course, it is necessary to notice that work-related computer use does serve only a limited range of occupations (Blom et. al. 2002, 6-7). For example, most traditional working-class occupations, such as plumbers or carpenters, have remained rather untouched with the use of computers in their jobs. But how much time was spent on computing in the Finnish households in the late 1980s and 1990s? There has been a notable increase in the average time spent on computing during in ten years; from less than 10 minutes a day to over 20 minutes a day. However, this finding is not an exceptional one if we consider the 18 The following analysis is restricted to those respondents that have computer access at home (1987/1988 N=2631; 1999/2000 N=4797). 155 development of the processor speed of computers and the increase in the supply of software after the late 1980s. For example, graphical user interfaces were used only in the Macintosh computers in the 1980s. Also the Internet was unavailable before the 1990s. These kinds of changes have probably influenced the primary purposes of home computers. In the 1980s computers were used for a very limited range of purposes. Computing meant ‘serious’ computing such as programming and word processing or just simple gaming. By the late 1990s computers have become almost ‘basic tools’ of work and entertainment for many households. The use of computer can range from Internet browsing to tutorial language lessons, for example. However, our analysis is confined to the examination of time spent on computing in general.19 Table 2 shows the average time spent on computing by socio-economic characteristics and changes by year. At first look, it appears that the differences in the time use on computing are rather similar to differences in the possession of computers. Students spent more time on computing than respondents in other labour market positions in both years. However, those in the position labelled as ‘other’ spent the least time in 1987/1988, but in 1999/2000 those who perform domestic work appeared to spend the least time with computers. Unemployed did not use time on computing at all in the late 1980s but spent over 14 minutes a day in the late 1990s. Time use patterns by socio-economic position are also interesting. All socio-economic groups have increased their computer use since the late 1980s, except farmers. Socio-economic group that spent the most on computing in both years are manual workers. This is a rather surprising finding because the possession of home computers was not very common in ‘worker’ households in the 1980s or 1990s. In this sense, it can be considered that there has been distinct ‘democratisation’ of computing, which cannot be observed as clearly in the diffusion of home computers into households. 19 One of the main weaknesses related this study is that time use on computing is not at all disaggregated in the Finnish Time Use Survey 1987/1988. This means that we are able to know only the total time spent on computing, but not what was done with computers. More itemised time use categories (such as playing games, word processing, Internet browsing etc.) would thus be needed for making adequate comparisons. The use of the Internet, however, is represented as a distinct category, but only in the 1999/2000 data set. It should also be stressed that no weight coefficients, indicating how many cases out of the total population are represented by a particular case, were applied in the analysis procedure. 156 Table 2 Time use on computing in Finland by labour market position, socio- economic status, age, gender and level of education in the late 1980s and 1990s (mins per day) 1987-1988 1999-2000 Change N Labour market position Employed 3,2 9,0 +5,8 1476/3381 Unemployed 0,0 14,2 +14,2 28/228 Homemaker 0,5 5,2 +4,7 38/139 Other 2,7 18,6 +15,9 64/335 Student 20,1 45,0 +34,9 1025/1711 Socio-economic status Farmer 4,6 2,8 -1,8 41/172 Other entrepreneur 3,8 9,1 +3,3 171/238 Professional, manager 2,0 8,4 +6,4 366/946 Intermediate non-manual 1,3 8,2 +6,9 514/1140 Manual worker 5,6 11,5 +5,9 464/778 Age 10-17 years 23,2 52,0 +28,8 905/1149 18-30 years 6,2 24,0 +17,8 418/1221 31-45 years 1,5 8,5 +7,0 952/1690 46-50 years 1,3 6,5 +5,2 324/1464 Over 60 years 2,2 16,5 +14,3 32/269 Gender Male 15,8 31,9 +16,1 1527/2806 Female 1,3 9,4 +8,1 1104/2988 Level of education Lower secondary 3,0 13,8 +11,8 840/1902 Lowest upper level 0,6 9,7 +9,1 127/895 Bachelor 0,9 13,9 +13,0 70/416 Master 2,2 9,7 +7,5 135/435 Ph.D. 14,7 4,6 -10,1 15/50 Total 9,71 20,26 +10,55 2631/4797 Differences by age, gender and level of education show relatively expected patterns of time use. In both years, the youngest age group spent most time on computing. The average time spent on computing has increased also in other age groups, but not as much as in the youngest one. Both men and women use more computers in the late 1990s than in the late 1980s. However, men have increased their use clearly more than women. Differences between educational groups produced interesting results. Respondents with academic post-graduate degree report a ten-minute decline in the average time use between 1987/1988 157 and 1999/2000. This is a rather strange finding since all other categories report considerable increase. However, the finding is probably due to the fact that there are very few cases in this group in 1999/2000 data (N=15). The previous findings indicate that socio-economic differences in computing have converged in Finland since the late 1980s. The average time use on computing has increased quite much in all the examined social groups; only farmers and people with post-graduate academic degree spent less time on computing in 1999/2000 than in 1987/1988. These findings suggest that computing has become a common time use category, which relates to the daily routines of most people. It has also become a category that can include time use for several leisure and work-related activities. This is because the basic characteristics of computing (i.e. what is actually done with computers) have probably changed quite much since the late 1980s. Particularly the role of the Internet should be considered in this respect because Internet has penetrated the average households in the late 1990s. It is reasonable to assume, for example, that the Internet use affects also other communication activities more profoundly than other types of consumer technologies. These kinds of suggestions have been put forward in recent American studies (see Robinson et. al. 2003). 8.5 Has computing become a ubiquitous activity? According to time use researchers, our ‘time budget’, that is, how we spend time, is an important determinant of our life changes (e.g. Gershuny 2000; Robinson – Goodbey 1999). It is also believed that the differentiation of individuals’ time use is considerably influenced by various economic and socio-demographic factors, such as income, class, age and gender. Many factors, ranging from disposable income to educational qualifications and gender roles, may lead to systematic time use differences among people. In this sense, it can be proposed that systematic exploration of individuals’ time budgets provides good possibilities to evaluate individuals’ positions in the social structure. Empirical findings presented in this paper show that there are structural differences both in the possession of home computers and time use on computing. These differences have become less clear by the late 1990s when computers became widespread in Finland. But considerable differences still exist. On the basis of our findings, we are able to observe a relatively clear socio-demographic pattern of computer use. Students and younger people spend more time on computing than any other social groups. On the other hand, however, the observed pattern does not seem to follow the basic social 158 and cultural hierarchies. For example, the possession of computers is most common in typical ‘middle class’ households but people in these positions do not spend much time on computing if compared to manual workers, for instance. It can thus be concluded that the differentiation of individuals’ time use on computing is influenced by socio-demographic factors, but the connections of these are to some extent multidimensional. In other words, the patterns of computer use differ rather irregularly between individuals belonging to different social positions. 159 References Bell, D. (1973) The Coming of Post-Industrial Society: A Venture in Social Forecasting. Basic Books: New York. Berkeley, G. E. (1971) The Administrative Revolution: Notes on the Passing of Organization Man. Prentice Hall: Englewood Gliffs. Blom, R. – H. Melin – P. Pyöriä (2002) Social Contradictions in Informational Capitalism: The Case of Finnish Wage Earners and Their Labour Market Situation. The Information Society, Vol. 18, No. 1, 1–11. Bourdieu, P. (1984) Distinction. A Social Critique of the Judgement of Taste (translated by R. Nice). Routledge & Kegan Paul: London. Castells, M. – Himanen, P. (2002) The Finnish Model of Information Society. Oxford University Press: Oxford. Dickson, P. R. (2000) Understand the Trade Winds. The Global Evolution of Production, Consumpiton, and the Internet. Journal of Consumer Research, Vol. 27, No. 2, 115–122. Florida, R. (2002) The Rise of the Creative Class. New York: Basic Books. Franzen, A. (2000) Does the Internet Make us Lonely? European Sociological Review, Vol. 16, No. 4, 427–438. Gershuny, J. (2000) Changing Times. Work and Leisure in Postindustrial Society. Oxford University Press: Oxford. Fidler, R. (1997) Mediamorphosis. Understanding New Media. Pine Forge: Thousand Oaks, California. Kraaykamp, G. (2002) Cumulative Advantages and Inequality in Lifestyle. A Dutch Description of Distinction in Taste. The Netherlands Journal of Social Sciences, Vol. 38, No. 2, 121–143. Lewin, C. – M. Orleans (2000) The Class Situation of Information Specialists: A Case Analysis. Sociological Research Online 5:3. , retrieved in August 2003. Lievrouw, L. A. (2001) New Media and the ‘Pluralization of Life-Worlds’. A Role for Information in Social Differentiation. New Media & Society, Vol. 3, No. 1, 7–28. Niemi, I. – H. Pääkkönen (1990) Time Use Changes in Finland in the 1980s. Studies 174. Central Statistical Office of Finland: Helsinki. Niemi, I. – H. Pääkkönen (2002). Time Use Changes in Finland through the 1990s. Culture and the Media 2002:2. Statistics Finland: Helsinki. 160 Nordic Information Society Statistics (2002) Nordic Information Society Statistics 2002. Nordic Council of Ministries: Helsinki. Nurmela, J. (1997) Suomalaiset ja uusi tietotekniikka. Katsauksia 1997/7. Tilastokeskus: Helsinki. (Finns and the New Technology) Nurmela, J. – H. Heinonen – P. Ollila – V. Virtanen (2000) Matkapuhelin ja tietokone suomalaisten arjessa. Katsauksia 2000/2. Tilastokeskus: Helsinki. (Mobile Phone and Computer in Everyday Life in Finland) Raijas, A. (2001) Kokemattomien käyttäjien näkemyksiä Internetistä ja sen tarjoamista palveluista. Elektronisen Kaupan Instituutin Julkaisuja. LTT – Tutkimus Oy: Helsinki. (Experiences about Internet Among the Inexperienced Users) Robinson, J. P. – G. Goodbey (1999). Time for Life. The Surprising Ways Americans Use Their Time. 2nd Edition. The Pennsylvania State University Press: Pennsylvania. Robinson, J. P. – M. Kestnbaum – A. Neustadtl – A. Lalvarez (2003) Information Technology and Functional Time Displacement. A paper presented at the 25th International Association for Time Use Research Conference in Brussels, Belgium, 17–19 September 2003. Rogers, E. (1986) Communication Technology: The New Media in Society. The Free Press: New York. Räsänen, P. (2003) Sosiokulttuurinen näkökulma: ovatko mobiili- uutispalveluiden lisäarvopalvelut tarpeellisia? Futura, Vol. 22, No. 1, 55–82. (A Socio-Cultural Perspective: Are the Additional Features for Mobile News Services Perceived as Useful?) Viherä, M.-J. – J. Nurmela (2001). Communication Capability as an Intrinsic Determinant for infomation Age. Futures, Vol. 33, No. 3–4, 245– 265. 161 9 POLARIZATION IN TIME USE? SOCIO- DEMOGRAPHIC BACKGROUND OF THE INCREASE IN TELEVISION WATCHING IN FINLAND Timo Toivonen Abstract Time used to television watching has increased in various countries. The phenomenon is scrutinized here on the basis of time use survey data of Finland from 1987/88 and 1999/00. Two alternative hypotheses were argued. According the first one differences between socio-demographic groups were converging in television watching, and according to the second one increase in watching was due to the divergence. Results showed that there were converging trends but most seeming were diverging trends, which indicated that television watching was growing strongly among people, who can be considered to some amount excluded from society, for instance unemployed men and unemployed young people. KEYWORDS: Reading, socio-demographic background, time use, TV watching 9.1 Introduction According to the Jonathan Gershuny’s seminal book (2000) differences in time use by social classes, age groups, genders etc. have been converging in developed countries over the period from 1960s to 1990s. Supposedly, this has been an indicator and reflection of the increase of equality in those societies. However, in the last years also in Nordic welfare countries, like Finland, have experienced changes. For instance, income differences have increased, and more and more people are excluded as unemployed and otherwise from the society. Also material living surroundings of people (for instance new media technology) have changed so much that one may ask, if the converging trends in time use have remained untouched. 162 9.2 Theoretical background In an earlier study of the present author it was found that differences in consumption expenditure structures between social strata diminished between 1955 and 1985. For instance, 37 % of all expenditure in 1955 was devoted to food and non-alcoholic beverage in working class families. In service class the corresponding figure was 30 %. The proportions of food fell in both group over thirty years, but more among workers, and corresponding figures were 21 and 18. An exception against these converging trends was expenditure devoted to education and culture (books, newspapers, attendance to theatre, concerts, movies etc.). Differences in percentages were growing. An explanation to this was the growth of entertainment electronics. In 1955 the entertainment electronics consisted in Finland practically only of radio, record player, and records. The spreading of televisions was only in the beginning. Therefore, in the earlier period people had no other alternative than read, go to movies, theatre etc. but in latter year 1985 people had televisions, videos, electronic games. In addition, acquiring car was possible to the vast majority of people. Therefore also the class differences became more visible in expenditure to culture. People had more alternatives in discretionary consumption. (Toivonen 1992.) It has been found that there have been exceptions to converging trends also in time use. One of them is television watching, where the differences between educational groups have increased: time devoted to television has increased in low level educational groups more than in high level educational groups (Gershuny 2000, 209—210)20. However, television watching time has grown commonly, i.e. the time used to television watching has increased also in upper layers of society (e.g. Gershuny 2000, and Robinson & Godbey 1997). Increase in time devoted to television is one of the most radical changes in people’s time use. According to the time use study in Finland the average time people devoted to television watching per day was in 1987/8 1 hour 45 minutes and in 1999/00 as much as 2 hours and 16 minutes (Niemi & Pääkkönen 2002, 64). This fact has been verified also on the basis of other studies (e.g. Pihanurmi 1998). This is the starting point of the study, and the research question is: Was the increase of time devoted to television watching approximately same in all socio-demographic groups or was the increase uneven between socio-demographic groups? 20 In fact, in Gershuny’s tables television watching and radio listening were combined, but on the basis of Finnish time use studies time used to radio listening is quite marginal in comparison with time used to television watching. 163 Traditionally television watching has not been a high-esteemed way to use leisure time. According to Frank and Greenberg (1980, 21) many have argued television watching to be a wasteland. Supposedly, attitudes have changed during the last decades, but still for instance Gershuny classifies television watching as “passive leisure at home” (2000, 207), and the attitude of late Bourdieu was stark: television is a threat to intellectual knowledge production (Bourdieu 1999). If this is true, we can easily explain the diverging trends in the spirit of Bourdieu (1984). People with high levels of education, occupational prestige, and income come to reject cultural forms that are popular among people of lower status. One of these popular forms is without doubt television watching. However, if also upper classes have increased their watching time, the Bourdieu-type explanation does not necessary come to difficulties, because according to Bourdieu-type explanation television watching should only increase less among upper social groups than among other social groups. It is the growing distinction, what is important! If upper social groups have increased their watching time, it is without doubt due to the fact that the television environment has changed radically. The increase in the number of channels has been enormous and the supply of television programmes has increased considerably during the last decade (Finnish Mass Media 2000, 47—51). The programme supply is diversified, and people can learn from television all kinds of new things on nature and society, enjoy highbrow culture etc. For instance, it has been found that in the USA already between 1965 and late 1970s more educated males increased their television watching time, when colour TV and more diverse programming were introduced (Juster & Stafford 1991, 501). Therefore, it is possible that in 1990s diverging tendencies in television watching between socio-economic groups were not more dominant tendencies. Thus converging trends are also possible, but neglecting the idea on television watching as passive leisure time use would not be wise, either. There is no reason to doubt that the result of an old 12-nation study is still relevant: television non-viewers rely more on alternative mass medias (like radio and newspapers), personal companionship, participate more also in formal social organizations, attended more movies, theatres etc. than television viewers (Robinson 1972). These are activities supposed to be found among economically active – not too busy – well educated people. Is it not reasonable to suppose that those people, who attend movies, theatre, opera etc. are more active than those people, who follow corresponding programmes in television? Therefore, we can consider that in generally television watching is more passive leisure time use than some other action. There are also things that either constrain or promote television watching. On the one hand, people, who are very busy in social rat race, have no time to 164 watch television. On the other hand, there has been a growing number of unemployed and old people, who are not practically able to do anything else than watch television. Also the increased supply of television programmes has an impact that those people are now able to watch television although around the clock. According to the above argumentation two alternative hypotheses can be presented: a) Because programme supply has increased and has become more diversified, those people especially, who have previously used their time to television watching less, like upper layers of society, have increased their television watching more than people on the average. Thus we can hypothesize that differences in time devoted to television were converging. b) Inequality in employment has increased. There are more and more very busy people, who have less and less time to watch television, but there are also more and more excluded people, who have lot of leisure time. Unemployment is higher among less educated than more educated. Therefore, we hypothesize television watching to have increased more among economically passive and less educated people than among economically active and well educated people. Thus we can hypothesise that differences in time devoted to television were diverging. In that case, we interpret that television watching is also a trait of passive leisure time. 9.3 Data and variables This study was based on the Finnish Time Use Surveys covering population aged 10 or over. Sample was stratified according to region, gender, and age, and the data collection period was from the beginning of April 1987 to the end of March 1988. Respondents were asked to fill in a diary for two days (one weekday, one weekend day) running. They were asked to record, in their own words, their primary activity, and what else they were doing at the same time. Record keeping was on a 10 minute-basis (Niemi & Pääkkönen 1990, 11—12, 97—101). The number of cases (days investigated) was approximately 15 400 in 1987/88 over the whole year. The weakness from the point of view of the idea of this study was the fact that time used to television watching was not at all disaggregated. We know only the total watching time of a person but not, what was used, for instance, to news, sports, or soap operas. 165 Data of the study of 1999—2000 was collected essentially in the same way, but the sample design was a little different from the design of the study of 1987—1988. The earlier study was based on stratified random sample. In the later study there was two phases. In the first phase the random sample from persons living in Finland aged 15 and over was drawn. In the second phase also all other persons, which were at least 10 years old and belonging to this selected person’s household, were included to the final sample. The collection was completed over the period between 1 March 1999 and 12 March 2001. (Niemi & Pääkkönen 2002, 111). The number of cases (days) was 10 500. These two data material were merged, and the number of cases in merged data was over 25 900. In practice, this number of cases was usable only in few analyses because of incomplete questionnaires. 21 The differences in the collection method mean that the two data are not completely comparable, because in the latter period television watching can not be considered as an individual independent action. However, it has been observed that television watching is in families often a social action (Morley 1986) or all the members of a family prefer same types of programmes (Barwise & Ehrenburg 1988, 56). It is also possible to speculate, what happens if tastes are different and the number of television screens is limited to only one or few in a family. Does this situation have such an impact that the persons with minority taste do not watch television at all? In any case, the behaviour of the members of a household is not independent from each other. Time used was explained by six variables in the study. First of them was gender. In the latest Finnish time use study it is reported that the difference between men and women in television watching increased (in the favour of men) in 1990s and an increase was evident already in 1980s (Niemi & Pääkkönen 2002, 37). However, results concerning only Finland are not consistent. In a study conducted 1994 it was not found any differences between men and women in watching time (Jääsaari & Sarkkinen 1995). Because this study is based on the same data as the study of Niemi & Pääkkönen, it would seem unnecessary to include gender in this study. However it was included because, firstly, we are interested relative differences in both periods, and secondly, gender is an interesting variable in the context and interaction with some other variables. Stage in family cycle is also a necessary variable, when the changes in time differences are to be explained. For instance, it can be assumed that lone providers have less time to television 21 Unfortunately, the time use study from 1979 (see Niemi 1983) could not be used here, because it was restricted only to September-November 1979. If these months would have been included from the latter studies, then the number of cases would have been so small that multivariate methods and multidimensional tables could not have been applied in analysis. 166 watching than singles in their own dwelling. Age is also an important variable because in earlier studies it has been found that older people watch more television than younger people. In addition, television watching in different cohorts, i.e. peoples born in the same period, was also inspected. Above mentioned variables can be connected with an life situation only in an indirect way. Instead we can assume that three other variables have direct connections. These variables were education (low), economic activity (unemployed), and income quintile (low), to which a person belongs by personal taxable income. By “indirect” we mean, for instance, that gender as such is not a sign of exclusion, but when it is connected with some other variable like unemployment, it can make situation of an unemployed better or worse. 9.4 Results The bivariate divisions of independent variables and television watching are documented in Table 1. According to these figures there were considerable differences between socio-demographic groups in television watching both in 1987/88 and 1999/00. For instance, unemployed people watched television 3 hours 17 minutes in 1999/00 whereas gainfully employed only 2 hours, respectively. Time used to television watching grew in almost every socio- demographic group, but there were clear differences in growth rates. Results support both alternatives of our hypotheses. For instance, time used to television watching has increased more in the groups with bachelor or master education (74 % and 52 %) than in the group of lower secondary level of education (47 %), where the time used was highest of all educational groups in 1987/88. True, there was decrease in the group of PhD or corresponding education, which is a contradictory result. The number of cases (108) in this group was low in comparison with other educational groups in this study, but absolutely it is not low, and points out the distinction making in the spirit of Bourdieu. In respect of age groups, the fastest growth has happened in the youngest groups, where the time used to television watching was lowest in 1987/1988. 167 Table 1. Time used to television watching in 1988 and 2000 by gender, stage in family cycle, education, economical activity, age, and income quintile (hours.minutes per day). 1988 2000 change % n Gender men 1.44 2.35 49 12 315 women 1.29 2.09 41 13 598 Stage in family cycle single living with parents 1.34 2.29 59 5 958 single in own dwelling 1.41 2.36 54 3 895 couple without children 1.50 2.32 38 8 382 couple with children 1.24 1.54 35 7 183 lone provider 1.23 1.54 37 331 Education lower secondary 1.31 2.24 47 9 485 lowest upper 1.20 2.04 56 1 937 Bachelor 1.12 2.05 74 842 Master 1.13 1.51 52 1 050 PhD 1.45 1.37 -8 108 Economic activity gainfully employed 1.27 2.00 38 14 720 unemployed 2.04 3.17 59 894 student and pupil 1.31 2.24 58 5 315 homemakers 2.11 2.55 22 706 other 2.26 3.03 34 4 278 Age class 10—22 1.30 2.24 61 6 100 23—35 1.25 2.10 53 5 655 36—48 1.30 1.59 32 6 203 49—61 1.47 2.21 32 4 676 62—74 2.19 2.59 29 2 400 75- 1.37 3.06 44 879 Income quintile lowest 1.34 2.28 57 5 148 2 1.45 2.36 49 5 147 3 1.39 2.39 61 5 147 4 1.33 2.09 39 5 147 highest 1.30 2.03 35 5 147 168 However, at least three observations refer to divergence, i.e. to the fact that differences between socio-demographic groups have increased. Firstly, in 1987/88 men watched television 15 minutes more than women (1.44 and 1.29). The growth in time used to television was among men faster (49 %) than among women (41 %), and thus the difference in watching time was already 26 minutes between genders in 1999/00. Secondly, the differences in television watching were quite equal between income quintiles in 1987/88 (personal incomes), but differences grew considerably during 12 years: the most intense growth happened in the lowest income quintile (57 %), and the lowest one in the highest income quintile (35 %). There are not many studies on television watching in different income classes, but according to these few studies there have not been differences between income classes (Jackson- Beeck & Robinson 1981). And thirdly, an observation that mostly supports the divergence hypothesis, is that according to economic activity, the time devoted to television grew mostly in the group of unemployed. Age was classified 10—22, 23—35 …75-, because this classification made it possible to inspect the behaviour changes of cohorts. Those, who were 10— 22 years old in 1987/88 were 23—35 old in 1999/00, those, who were 23—35 years old in 1987/88, were 36—48 years old in 1999/00 etc. According to the cohort theory, every cohort has “key experiences” in its youth. These experiences have impacts on the whole rest of life of that cohort. It is thus called “cohort effect” (Toivonen 1999, 269—298). In this context it means that every cohort learns in its youth to devote to television watching a certain time. The time continues to be approximately same during their whole life. This kind of observation is not supported on the basis of data. On the contrary, the time used to television watching grew remarkably in every cohort. In the youngest cohort (10—22 years old in 1987/8) it grew from 1 hour 30 minutes to 2 hours 10 minutes, in the next cohort from 1 hour 25 minutes to 1 hour 59 minutes and so on. Thus, we can not speak about cohort effect in a sense that the people’s television watching time remains same from one period to another. In the next step of the study the general multivariate linear models (in this case, Analysis of Variance, ANOVA) were applied to explain the changes in television watching. The strategy was as follows. The model (1) is called as basic or base model. It represents the impact of period on watching time. F- value was significant, which is quite natural because the sample was large. The explanation percentage was 4.5. It is high considering that individual variations in watching time were big in both periods: some people do not watch at all television, some people watch several hours per day. In 1999/00 13.7 % of all people reported that they did not watch television at all, and on the other side 2.6 % of people reported to watch at least seven hours per day. 169 Table 2. Television watching explained by background variables (Univariate ANOVA). df F Sig. partial ETA sqd Corrected model (1) 1 1214 .000 .045 Intercept 1 34303 .000 .570 Year 1 1214 .000 .045 Adjusted 100R2=4,5 Corrected model (2) 45 36 .000 .109 Intercept 1 1558 .000 .105 Year 1 38 .000 .003 Year*Gender 2 128 .000 .019 Year*SFC 8 10 .000 .006 Year*Education 8 5 .000 .003 Year*EA 8 30 .000 .018 Year*Age 10 4 .000 .003 Year*IncQ 8 3 .000 .002 Adjusted 100R2=10,6 Corrected model (3) 69 25 .000 .113 Intercept 1 384 .000 .028 Year 1 20 .000 .002 Year*Gender 2 32 .000 .005 Year*SFC 8 10 .000 .006 Year*Education 8 4 .000 .002 Year*EA 8 30 .000 .018 Year*Age 2 2 .125 .000 Year*IncQ 8 3 .003 .002 Year*Gender*Ed 8 1 .657 .000 Year*Gender*EA 8 5 .000 .003 Year*EA*Age 8 2 .015 .001 Year*Education*Age 8 1 .262 .001 Adjusted 100R2=10.9 Abbreviations: SFC = Stage in Family Cycle, EA = Economic Activity, IncQ = Income Quintile, Ed = Education 170 In the model (2) gender, stage in the family life cycle, education, economic activity, age, and income quintile interacted with period and television watching time (the first-level interactions). Interaction terms indicate, whether the changes between socio-demographic groups have happened at the same pace or not. For instance, if the interaction of time and gender is not statistically significant, then the change has happened at the same pace in both gender groups. Actually, the F-value of this interaction was highly significant and therefore the change did not happen at the same pace, which could be presumed already on the basis of Table 1. In fact, all interaction terms were significant, which means that practically in different categories of every socio- demographic group changes were not identical. Explanation percentage of the model (2) was 10.6, which indicates socio-demographic variables adding the explanation percentage considerably in comparison with the basic model. The best predictors in model (2) were gender and economic activity, because their partial ETA-coefficients were highest. The next phase in the analysis was to investigate the impact of the higher order or the second-level interactions. In model (3) four second-level interaction terms (four variables in term) were added to model (2). The purpose was to get more detailed information on the relation of independent variables and television watching. For instance, in discussions about excluded persons, one of typical theses has been a supposition that excluded persons are especially less educated men. Therefore, the first second-level interaction term was the interaction between television watching, year, gender and education. If television watching was growing disproportionally faster among the above mentioned group, we can carefully assume this to be evidence on increasing passivity of this group also in leisure time use in comparison with other groups. The second new variable was the interaction between television watching, year, gender and economic activity. The idea was same as in the previous case. Was the television watching growing disproportionally faster among unemployed men than in other groups? The third term was interaction between television watching, year, economic activity, and age. The group under scrutiny is in this case the group of young unemployed. The fourth new term was interaction between television watching, year, education, and age., in which case our attention is especially paid to the group of the young less educated people. The motivation to take the last mentioned term to the model is in the classical assumption of growing meritocracy in developed societies (Young 1963). It means that in a developed society, like Finland, every person, who has at least some capacity to learn is educated as long as possible. Thus, those, who have only minimum education, are passive and incapable people. However, in the less developed phase of these societies this kind of 171 massive education was not possible, and therefore in the less educated group of older people are persons, who are mentally as capable and active people as better educated younger persons, but who were not fortunate to get education. The new terms did not add much the explanation power, which is in model (3) 10.9 %. In the model (3) the term year*age is no more significant, but the lack of significance can be due for technical reasons.22 The interaction between television watching, year, gender, and education was statistically non- significant as well as the interaction between television watching, year, education and age. Thus the assumptions presented in the previous paragraph on interaction between education and age and education and gender were not tenable. Instead, the interaction between television watching, year, gender, and economic activity was significant, as well as interaction between television watching, year, economic activity, and age. However these parameters do not uncover interactions in detail, i.e. what are figures in different cells. 22 The model (3) could not be identified with the original age classification (too many empty cells). Therefore age classification was dichotomous: people under 36 years old, and people over 35 years old. Thus the terms of model (2) and model (3) including variable “age” were not identical, although the outlook is. 172 Table 3. Means of time used to television watching by year, economic activity, and gender (hours.minutes per day). Economic Activity Gender Time Change % n 1988 2000 Gainfully employed Male 1.35 2.15 42 7 454 Female 1.19 1.46 34 7 266 Unemployed Male 2.21 3.51 65 443 Female 1.44 2.47 61 451 Student and pupil Male 1.39 2.31 52 2 495 Female 1.24 2.19 65 2 820 Other Male23 2.26 3.20 36 1 923 Female 2.01 2.38 31 3 061 Therefore, the Table 3 on means of time used to television watching by year, economic activity, and gender was produced as well as Table 4 on means of time used to television watching by year, economic activity and age. Many interesting observations can be made. Firstly, separately, both in female and male unemployed groups the growth in watching time were faster (65 % and 61 %) than it was in total group of unemployed (59 %, Table 1). This is due to the structural change. The increase of female unemployed – who watched television less than male unemployed – was faster than that of male. Secondly, it is easy to see that television watching did not grow consistently in every group of economic activity more among men than women. Increase in time devoted to television was stronger among female students than male students, and it was quite equal both in female and male employed. These observations point to convergence between genders but growing divergence between employed and unemployed. 23 Homemakers and others have been merged because the number of male homemakers was only 26. 173 Table 4. Means of time used to television watching by year, economic activity, and age group (hours.minutes) per day). Economic Activity Age Time Change % n 1988 2000 Gainfully employed -35 1.22 2.04 51 5 627 36+ 1.31 1.59 31 9 093 Unemployed -35 1.56 3.32 83 402 36+ 2.18 3.00 38 492 Student and pupil -35 1.32 2.25 58 5 203 36+ 1.28 2.04 41 112 Homemaker -35 1.25 1.45 23 379 36+ 1.47 2.23 34 327 Other -35 2.09 2.36 21 144 36+ 2.17 3.04 34 4 134 All EA groups -35 1.28 2.18 56 11 755 36+ 1.45 2.23 36 14 158 Time used to television watching rose among the younger group 56 % and among the older one 36 %, and the younger group almost reached the older one in television watching (Table 4). However, the most important observation in table 4 is that the increase in watching time was among young unemployed much faster than among old unemployed. In the former group increase was 83 % and in the latter one 38 %, which was not much more than average rise in the older age group (36 %). The growth of watching time of younger unemployed was so strong that they watched television in 1999/00 more (3.32) than the older unemployed (3.00), although the situation was reverse in 1987/88. It is interesting that among homemakers, who were over 35 years old, increased their watching time more than under 36 years old. Evidently, younger homemakers could not increase much of their watching time because of small children? 174 Table 5 Means of time used to reading by year, economic activity, and age group (hours.minutes per day). Economic Activity Time age 1988 2000 change % n Gainfully employed -35 .40 .28 -31 5 627 36+ .51 .45 -12 9 093 Unemployed -35 1.09 .33 -52 402 36+ 1.03 1.02 -2 492 Student and pupil -35 .45 .30 -33 5 203 36+ 1.20 .40 -50 112 Homemaker -35 .43 .26 -31 379 36+ 1.06 .48 -27 327 Other -35 1.00 .52 -13 144 36+ 1.23 1.10 -16 4 134 All EA groups -35 .43 .30 -32 11 755 36+ 1.01 .53 -13 14 158 Tables 1–4 indicate perhaps more on divergence than convergence between socio-demographic groups in television watching: economically passive have become more passive in their leisure than economically active. Or was it so? Reading is considered without doubt as an active leisure time use at home. Therefore, passivity thesis was controlled by producing Table 5, which is otherwise the same as Table 4, except that the dependent variable was time used to reading (books, periodicals, newspapers etc.). The time devoted to reading decreased in all categories of age and economic activity but more in the younger group (-32 %) than in older one (- 13 %). Differences between age groups were in the same scale also in all other groups of economic activity except in the group of unemployed. In this group the development of younger and older age groups was strongly diverging. In the older group there was practically no change in respect to time devoted reading (-2 %) but in the younger age group the decrease was dramatic or -52 %. This means that during 1990s not only passive leisure increased among 175 economically passive more than among economically active, but also active leisure decreased more among economically passive than economically active. 9.5 Summary and discussion During 1990s time used to television watching increased in Finland very strongly. The purpose of this study was to find out possible differences of this growth between different socio-demographic groups. The phenomenon was scrutinized here on the basis of time use survey data of Finland from 1987/88 and 1999/00. Two alternative hypotheses were argued. According to the first alternative the differences between socio-demographic groups were converging in television watching, because the supply has increased and a lot of educational programmes, qualified films etc. - esteemed by better educated higher layers of society - are available. According to the second alternative of the hypothesis watching was growing due to the divergence. Some socio- economic groups increased their television watching more than some others. Results show that there were converging trends. For instance, differences between age groups were levelling out (young people come in the TV watching time closer to older people). However, most perceptible were trends, which indicated that passive television watching was growing strongly among people, who can be considered to some amount in underdog groups of society, for instance, among unemployed men and unemployed young people. This increasing passivity was supported also by observation that time devoted to reading was decreasing most strongly among the group of young employed. Thus, it seems that the inequality in the Finnish society was growing during 1990s not only in material but also in intellectual sense. However, we can not be sure about the increasing passivity on the basis of time use surveys, because in these surveys time use has not been disaggregated to programme types. As programme supply in television has increased and diversified (or has it?), it is possible that higher layers of society increased more their television watching than lower ones. This study does not support this hypothesis. However, the future can make a change. For instance, travelling has become less convenient in modern times, on the one hand, because of congestions and delays and, on the other hand, because of wars, terror attacks, etc. One may ask, if one of the purposes of tourism is learning, as highly educated people often use to say, is it then much more easier to get acquainted with some exotic place by virtual tourism, by watching television, than by real tourism, by travelling to the actual place? 176 References Barwise, T.P. – Ehrenburerg, A.S.C. (1984) The Reach of TV Channels. International Journal of Research in Marketing, Vol. 1, No. 1, 37–49. Bourdieu, Pierre (1984) Distinction. A Social Critique of the Judgement of Taste. Harward University Press: Cambridge, MA. Bourdieu, Pierre (1999) Televisiosta. Otava: Helsinki. Finnish Mass Media (2000) Culture and Media 2000: 1. Statistics Finland: Helsinki. Frank, Ronald E. – Greenberg; Marshall g (1980) The Public’s Use of Television. Who Watches and Why. SAGE Publications: Beverly Hills. Gershuny, Jonathan (2000) Changing Times. Work and Leisure in Postindustrial Society. Oxford University Press: Oxford. Jackson-Beeck, Marilyn – Robinson, John P. (1981) Television Nonviewers: An Endangered Species? The Journal of Consumer Research, Vol. 7, No. 4, 356–359. Juster, Thomas F. – Stafford, Frank P. (1991) The Allocation of Time: Empirical Findings, Behavioral Models, and Problems of Measurement. Journal of Economic Litterature, Vol. 29, No. 2, 471–522. Jääsaari, Johanna – Sarkkinen, Raija (1995) Naiset ja miehet radionkuuntelijoina ja television katselijoina. Sana, Elina (toim.) Naiset, miehet ja uutiset. Yleisradion tasa-arvotoimikunnan julkaisuja. Sarja A:1. Yleisradio: Helsinki. Morley, David (1986) Family Television: Cultural Power and Domestic Leisure. Comedia: London. Niemi, Iiris (1983) The 1979 Time Use Study Method. Studies 91. Central Statistical Office of Finland: Helsinki. Niemi, Iiris – Pääkkönen, Hannu (1990) Time Use Changes in Finland in the 1980s. Central Statistical Office of Finland. Studies 174: Helsinki. Niemi, Iiris –Pääkkönen, Hannu (2002) Time Use Changes in Finland through the 1990s. Statistics Finland. Culture and the Media 2002:2: Helsinki. Pihanurmi, Vesa (1998) Television katselu vuonna 1997. Radio- ja TV- tutkimuksen vuosikirja 1998. Robinson, John P. (1972) Television’s Impact on Everyday Life: Some Cross- National Evidence. In: Rubinstein, Eli, Comstock, George & 177 Murray, John. Television and Social Behaviour 4. Government Printing Office: Washington. Robinson, John P. – Godbey, Geoffrey (1997) Time for Life. The Surprising Ways Americans Use Their Time. The Pennsylvania State University Press: Pennsylvania. Toivonen, Timo (1999) Empiirinen sosiaalitutkimus. Filosofia ja metodologia. WSOY: Helsinki. Toivonen, Timo (1992) The melting away of class differences? Consumption differences between employee groups in Finland 1955–1985. Social Indicators Research, Vol. 26, No. 3, 277-302. Young, Michael. (1963) The Rise of Meritocracy 1870–2033. Penguin: Harmondsworth. 178 179 10 ANALYZING CULTURAL TOURISM IN EU: YOUNG ADULTS’ CULTURAL ACTIVITY AND CONSUMPTION WHILE ON HOLIDAY∗ Taru Virtanen Abstract The paper scrutinizes the role of cultural capital (a concept introduced by late Pierre Bourdieu) in decision-making in cultural tourism as well as in travel patterns. In this article one form of cultural tourism, which could also be seen as an enlargement of the concept or notion of cultural hobby/ism is studied. Cultural tourism is thus considered to be something related to the culturally involved consumption while on holiday. Cultural capital is studied in light of the consumed cultural products and activities while on holiday. Cultural elements referred here can be classified as material and immaterial products of high culture, which can be seen in opposition to the objects of so-called popular culture. The target group of the study is the young Europeans, between 18 and 34 years of age. Examining the concept of cultural capital the length of education is essential as well. It seems that also regional differences in travel propensities and patterns emerge, at least when younger cohorts of European travelers are studied. Nordic youth seems to have their distinguished ways of traveling among other Europeans, and these features are brought out in this paper. Thus, also the Nordic aspect of tourism and its relation to cultural capital is to be emphasized. KEYWORDS: cultural tourism, cultural capital, young adults, education, EU ∗ An earlier version of this paper (“Young European Tourists and Their Consumption Motives – Does Cultural Capital Play a Focal Role? “) has been presented at the 6th Conference of the European Sociological Association (ESA), in Murcia, Spain, 23-26th September, 2003, in the Sociology of Consumption Network Stream. I would like to thank the network for active participation in the discussion on my paper. I would like to express my deepest gratitude to both Antti Honkanen and Timo Toivonen who have reviewed my paper carefully and given me most valuable advice regarding this article and my subsequent research alike. 180 10.1 Introduction Cultural tourism has been a popular and widely used concept and topic in both academic and more mundane writings for some time now. A besetting sin for a scholar is to refer too cursory to cultural tourism when various types and patterns of tourism are concerned, where as one should start by defining the boundaries and characteristics into which the certain type of traveling at certain point of examination fits. Culture can be seen anywhere where there has been an intervention of human civilization. On my opinion this is an anthropological outlook on culture. It means that a man’s hand has shaped nature to his own purposes (we can also see here a parallel to Marx’ view of a man and his creativity, see Bocock 1995, 47–48), was it a place of worship for ancient Gods or an aqueduct or other means of infrastructure or even an amusement park with it’s surreal surroundings (see ‘McDonaldization’ by Ritzer 2001, 181–202). If thus everything anthropology considers culture would be the interest of a cultural tourist, would the field of cultural tourism researcher be vast and never- ending. Cultural tourism is quite often seen to be in reference with culturally interesting monuments, and often value is set on the (old) age of object of interest. Artifacts of alien cultures are thus on the culture tourist’s list, and the abstract phenomena of culture is thus seen rather anthropological way. In this article, on the other hand, the attempt is to study one form of cultural tourism, which could also be seen as an enlargement of the concept or notion of cultural hobby/ism. Here, cultural tourism is thus considered to be something related to the culturally involved consumption while on holiday. The endeavor of this article is to scrutinize the travel patterns, which prevail in the ways of European young adults as far as cultural elements and their consumption is concerned. Cultural elements referred here can be classified as material and immaterial products of high culture. High culture and high consumption is thus seen in opposition to the objects of so-called popular culture. Important tool, and we could also say point of view, in this attempt is Bourdieu’s concept of cultural capital, which has occurred in some, but relatively few, previous studies on cultural tourism (Richards 1996). 10.2 Cultural Tourist Meets Cultural Capital in Social Context Despite the fact, that cultural capital has been since its birth operationalized, there are only few studies where this has been successfully applied. The purpose of this paper is to study the occurrence and use of the cultural capital in tourism, as it can be seen to be a way to discover the possible inequality and 181 distribution of opportunities that prevail in the society also more universally. Cultural capital has been adopted in previous studies mainly in order to elucidate the social stratification in consumption in general, but in the light of tourism consumption the concept of cultural capital and it’s empirical counterparts has been studied rather little. As Meetham (2001, 114) states, boundaries between high and low culture are dissolving, as tourists has started consume space as culture. By this definition one could characterize nearly anything as culture and tourism, since culture has penetrated into everyday life. Tourism and daily life of consumption has begun to melt as the tourist gaze diffuses the boundaries between the tourist and the non-tourist social activities (Urry 1996, 115–116). This can be seen for example in actions that imitates daily actions such as work – tourism has often been seen as stressing and scheduled as work, and not the way to relax and forget the limitative nature of time (Selänniemi 1997). Urry also brings the social inequality into tourist discourse in his theory of tourist gaze. This means that tourism is highly socially involved, and the essence of tourism is to purchase “a particular social experience” (Urry 1996). Tourism is thus socially stratified and within tourism we can find multiple sociologically interesting factors. We could further argue, that tourism research has a significant role, because of tourism’s fur-flung field, when explaining broader social phenomena of today’s society. This is why it is believed that through examining tourism consumption it is possible to acquire knowledge also on the phenomena of society on the whole. Culture can be defined in many ways. The most commonsense way to apprehend consumable culture products is to include high arts, literature, music, theatre and so, in the calculation. Even though being quite commonsense the division is still quite apt also in the scientific sense. These products also interconnect with the bourdieuan idea of superior knowledge being the most significant characteristic one needs in order to (here) consume in the right - in the light of the cultural capital - manner. Superior knowledge on something is pursuit and achieved through education and this, on the other hand, leads to social exclusivity, which is the primary aim or by-product of power struggle with the forms of capital. The more Bourdieuan way to express the superior knowledge would be referring to it as “good taste”. (Meetham 2001, 115; Mäkelä 1994.) According to recent studies on consumption and cultural capital, we have chosen so called highbrow culture as distinct from popular culture to be included in the concept of cultural capital (see e.g. Katz-Gerro – Shavit 1998). In Katz-Gerro and Shavit’s article education seems to be a push factor towards consumption of higher cultural products. The results still argue that despite 182 more educated people’s interest in highbrow culture, they did not lose their interest in popular cultural way of life either. 10.2.1 Good Taste and High Culture – The Formation of Cultural Capital Cultural capital is a concept, which requires both good apprehension of previous literature and sociological imagination in order to be measured and analyzed. In defining cultural capital Bourdieu has seen upbringing, influence and atmosphere of one’s family to be extremely crucial (Bourdieu 1994, 36– 37). Thus the level of education, or more simply time spent in educational institutions defines the cumulated amount of cultural capital. The formation of cultural capital is still not as simple as presented above. In order to scrutinize one’s power structure we need to go deeper into what Bourdieu meant by cultural capital. Bourdieu (2002) presents three different forms for cultural capital to exist: embodied state, objectified state and institutionalized state. These three represent the cultural aspects of society and ‘culture’ in terms of everything material and immaterial interfered or created by human being. Bourdieu sees cultural capital appearing in the works of art (objectified state), in education and it’s degrees (institutional state) and lastly in individuals in the form of habitus (embodied state). In measuring certain kind of consumption –such as going to museums– using educational level as one of the independent variables we can see Bourdieu’s idea also in this study, as we are interested in young individuals’ culturally involved tourism. According to national statistics (for example in Finland Tilastokeskus - Statistics Finland) as well the data in hand, people in the Scandinavian countries spend time in educational institutes as relatively old age compared to other students in the EU. The average age of finishing full-time education is 19,21 among young people (18 to 34 years) in all EU countries in total. The Nordic youth are 21,1 years old in average when finishing their schooling whereas the younger to swift to working life are the British and the Irish at the age of 17,7. (Central Europeans 18,77 and Mediterranean 19,49). This occurs also in the OECD statistics, as they state that in Nordic countries even as much as 30 percentages of the 30-year-old people are still students whereas the average for the OECD countries is 13 of the age group 25–29 year olds (Kyrö 2002). As a curiosity it can also be mentioned that according to the above mentioned statistics time spent in higher educational institutions is for example in Finland 4,1 years in average and the average of all OECD countries is two and a half years. This could suggest in bourdieuan light that 183 the Nordic youth and young adults possess rather large amount of cultural capital in comparison to its other European counterparts. 10.2.2 Research Questions and Data As stated earlier, in this article, one special form of cultural tourism is studied, which could also be seen as an enlargement of the concept or notion of cultural hobbyism. This relates to the idea, that an individual who is enthused by high consumption in general, quite likely is to reflect this hobby of his/her (referred here as ‘hobbyism’) to his/her traveling, too. Thus cultural tourism is considered in this context to be something related to the culturally involved consumption while on holiday. The patterns of cultural tourism will most likely to be quite multifaceted in each individual’s case. The purpose of this study is to find out the potential prevalence and/or use of cultural capital in tourism consumption context. The target group under scrutiny is young people under 35 years of age, and two subgroups was formed dividing the sample into ‘young adults’, aged from 18 to 25 years and ‘senior young adults’ (sic!) aged from 26 to 34 years. The research questions in this study are shortly as follows: • Can we find cultural capital having any role in tourism and culturally involved consumption while on the holiday? • How can we measure cultural capital in tourism context? • What kinds of culturally involved consumption patterns are predominant among the young EU citizens’ tourism? The answer to research questions is sought with the help of Eurobarometer data (Melich 1999), which covers 15 EU member nations in the year 1997 (a complete list of the variables used in the analysis can be found in the appendix, p. 211.). The survey deals with issues concerning tourism and the total number of cases is 16 168. The sample of young adults (18–34 years of age) is in the unweighted data 5 193, but in order to minimize the risk of too low cell frequencies weighted data was used. The number of the respondents in the weighted sample is then 50 906 155. Closer information on the weighting procedure can be found in the study carried out by Toivonen (2001, 43–44). 184 10.3 Conceptualization of Cultural Capital in Tourism – Variables Used in the Analysis In order to study cultural capital in tourism we first need to clarify the concept of tourism and the contents of cultural capital. Tourism is seen here as leisure activity. Tourism is thus action on voluntary basis, and one of the alternative ways of spending the residue time from work. Tourism can be seen as a hobby or something that should be a source of pleasure, escapism and relaxation. (Pronovost 1989, 56; Selänniemi 1997, 37) Tourism includes both traveling away from home or other daily surroundings for leisure purposes (in some studies also business trips are seen as tourism, but in this study the definition of ‘holiday’ and its relation to leisure depends on the respondent’s view), and as the long (vs. short) trips are in question staying in the destination(s) four nights or more is required. (WTO 2003) In the survey the most central question is about whether the respondent has traveled during the year in question (once or several times), or whether s/he is still planning but has not yet traveled, or will not travel at all. Tendency to tourism in general is thus measured with a dichotomous variable with alternatives ‘one or more trips’ or ‘no trips’ (where ‘still planning to travel’ is combined with ‘no trips that year’). We need to pay attention in the following analysis that the questions regarding traveling or holiday trips in the data refer both to domestic and international tourism. This notion is quite vital when discussing the emerging disparities in traveling inclination in general and tourism consumption patterns in particular among young adults from different parts of EU. 10.3.1 Cultural Consumption Variables Sanaksenaho (2002, 148–164) has used both cultural and social capital in her research on inequality in Finnish society (in general, not tourism) in Bourdieuan way. In her study, such consumption or leisure activities as going to theaters or concerts, listening classical music, reading literature, painting/drawing and playing an instrument represented cultural capital. Such less cultural hobbies as going to movies or restaurant or even church and working out were consider not to reflect the essential idea of the cultural capital, and were thus omitted from the analysis. Sanaksenaho justifies these choices on the basis of previous literature on the subject, such as Alasuutari for instance (1997). Katz-Gerro and Shavit have applied similar kind of criteria for representatives of cultural capital in their study on class structures in Israel (1998). 185 Due to the limitation our data sets us, the comparisons were done between and distinctions were searched in the consumption acts such as purchasing music and books, where as purchasing local crafts or food was seen not to represent culturally involved consumption. Other intermediates of distinction were events or sites visited during the holiday. The data puts forward the alternatives, which are fully listed in the appendix, and from which the most interesting ones for this study are: ‘Museums, exhibitions, archeological sites’, ‘Churches, temples, mosques, synagogues, etc.’, ‘Towns or areas (sites) of historical, cultural, or architectural interest’, and ‘concerts/music events’ (from which the bolded part is to be referred in the tables and figures subsequently). Traveling usually requires planning beforehand. The prospective tourist weights his/her desires and preferences trying to maximize the benefits s/he hopes to acquire from the holiday and from a particular destination. Simply, if one needs to relax and ‘chill out’, a vacation by the sea on the beach could be the answer. On the other hand, for the tourist who constantly is in quest of knowledge (on foreign cultures) and possesses the ‘needs’ of self-actualization and curiosity, we could see this traveler to be flying to, say, Rome. This division is a bit rough, but it should present the idea of the importance of choosing (the type of) the destination. The actual decision is also somewhat based on the season; we could predict the typical weather of a season playing significant role when choosing between a beach resort or a weekend in a buzzing city. The survey includes questions about the destination of the holiday, presenting the alternatives stated in the appendix, from which for better revision the alternatives ‘city’ and ‘sea’ were chosen. The comparison between city and sea destinations was seen interesting since we could decipher that there is difference in cultural involvement when it comes to choosing the type of holiday destination. In the questionnaire, the question regarding the destination referred to three longest holiday trips, and several types were possible for one individual trip (for example a trip to both sea and city destination). In this analysis, however, only the first trip (of 4 nights of more) was included. It seems that although three longest trips were included in the questionnaire and their type was asked, the vast majority of respondents had only had just one trip during the year, and thus, it was rather fruitless to study all the three trips separately and jointly. 186 10.3.2 Demographic Variables Essential to the accumulation and/or formation of cultural capital is also education and especially its amount or length. In this study education structure among the young Europeans was studied with the help of a variable ‘age of finishing full-time education’. This is quite simple but still relatively valid way of measuring the amount of potential cultural capital. Education structure is studied with the variable, which states the respondent’s age of finishing the full-time education, as mentioned above. Four categories constitute the education variable: ’up to 15 years’, ‘16 – 19 years’, ‘20+ years’ and ‘still studying’. When examining the younger cohort (18–25 years) it is obvious that the category ’20+’ is more infrequently mentioned as with the less young cohort. In order to make better comparisons, the countries of origin were transformed into four regions, which could be seen internally quite homogeneous when it comes to the “culture” or way of life. This division is somewhat similar to Esping-Andersen’s division of regimes, which derives its base from the similar kinds of social structures of the societies belonging to a same region. The ground for division of countries has never been unambiguous, since, as always with classification, unique characteristics of countries tend to reduce and thus data is lost for some part. (Härkönen 2002, 34.) An adaptation of Esping-Andersen’s regime division was used in this analysis. In the original model there are only three regimes, in which the conservative regime includes most of the Central European countries and Mediterranean countries as well. In the light of tourism we can regard the latter ones being so distinguished from the previous, that it is more suitable to separate the Mediterranean countries into their own regime. The original labeling of the regime model (ibid.) was omitted as unsuitable or unnecessary for this particular study. Europe was thus divided geographically forming the regions of ‘Nordic’ (including Finland, Denmark and Sweden), ‘Central Europe’ (Germany, Austria, Benelux and France), ‘Mediterranean’ (Italy, Greece, Spain and Portugal) and lastly ‘UK and Ireland’. These 15 countries were EU members on the year of data collection (1997), and subsequently whenever referred to “Europe” or “Europeans” in this study, these 15 countries are valid. As Wilska (1999, 13) states, cultural and postmodern aspects are focal in studying consumption but nevertheless also the importance of economic factors is significant. Money is the most important means of consumption and the presence or lack of it is substantially critical even though aspects such as taste, life style and identity has been prominent in examination of present 187 consumption patterns. The same applies also in the consumption of tourism (which will be referred subsequently simply as ‘tourism’ or ‘tourism activity’). Earlier studies show the significance of economic factors in comparison to aspects emphasized in postmodern discourse. Even though postmodern theories stress the insignificance of class and thus income or socio-economical status, it has still been shown in empirical studies that in the era of postmodern consumption the demographics do play a focal role. Income variable is studied implicitly in this paper through consumption while on holiday. Other central demographics that play a significant role in this paper are age and nationality (or better: region of residence). (Räsänen 2002; Toivonen 2000.) 10.4 Methodology In this paper tourism is studied in tandem with the concept cultural capital. The focus of the study could then be defined through this combination as cultural tourism. Cultural capital can be empirically measured arguably in various ways, but in this article a method based on previous literature was constructed, and the most inspiring source is Bourdieu (1984) as well as Katz- Gerro and Shavit (1998). In Finland Sanaksenaho (2002) has quite recently studied cultural capital in general in Finnish class structure by using cluster analysis. The linkage between these above-mentioned researches is a specified definition of highbrow cultural products and their interconnection with consumption and hierarchy in the society. This study could be characterized as an explorative one, and thus the methodology used gives us more like guidelines or trends as well as a possibility to deepen the perceptions on tourism interrelated with cultural capital. This study aims to link cultural capital with tourism consumption as well as elucidate the travel patterns of young European adults. In this light the cross-tabulation used can be seen well grounded. Cross-tabulation gives us clues but still does not justify us drawing conclusions on causal relations or their direction. In order to do so, we would need to deepen the analysis with the help of, for example, logistic regression, as dichotomous dependent variables are in question (Toivonen 1999, 319). 10.5 Empirical Findings on Young Adults’ Cultural Tourism Tourism is a popular way of spending leisure time within the young European adults. 54 per cent of young adults had been on at least one holiday trip during the year in question, and respectively 51 % in the younger group and 56 % in 188 the older one had been on holiday. The young adults’ inclination of taking holiday trips is seen in Figure 1. Figure shows the percentages of young adults who have taken at least one trip during the year in question by the region of residence, and the sample is divided into two subgroups according to age. One or more trips by region 77 59 56 68 46 57 45 5151 60 0,00 20,00 40,00 60,00 80,00 100,00 Nordi c Centr al E Medit errane an UK+Ir e Total % 18-25 26-34 Fig. 1. Trips (at least one) made by young Europeans according to their nationality (region), percentages. (n =50 212 085) As we can see, the Nordic youth, that is young adults from Denmark, Sweden and Finland, seem to be most eager ones to travel. Quite surprising is that Central Europeans do not travel as much as others, even though it would be expected, since their geographical vicinity or outreach to places could be seen to support traveling. The older age group travels more, which could be predicted to be caused by bigger income24. As we are interested in cultural capital and its role in tourism we add the demographic variable on education also in the array. Time spent in educational institutes refers to the potential of accumulate cultural capital, and thus we measure the relationship between taking holiday trips and the age of finishing full-time education. The results can be seen in Figure 2. 24 The income variable is not presented explicitly in this article, but for the reference we could mention that in comparison between the two age groups, without exception the older age groups in all regions have bigger share of the upper income classes than the younger counterpart. For example in the Nordic region 76 % of the younger group had income less than 2801 US dollars where as the older group had only 58 % correspondingly. (N of valid cases = 33 975 186; it must be taken into account that as much as 33 % of the respondents have not announced their income.) 189 One or more trips 35 55 66 44 35 55 63 51 60 56 0 20 40 60 80 Up to 15 yrs 16 – 19 yrs 20 + yrs Still studying Total Age of finishing education % 18-25 26-34 Fig. 2. Trips (at least one) made by young Europeans according to their age of finishing full-time education, percentages. (n =50 212 085) If education and cultural capital can be seen to have (causal) association, then we could argue that cultural capital has rather important role in taking trips. The tendency to travel grows linearly as the level of education rises, which means that the more educated individual in question, the most liable s/he is to travel. We must still take into account the economic factor again. As we know, traveling requires, as much as cultural capital (possibly), economic capital as well. We also intuitively know that education relates to better employment opportunities, which on the other hand links with higher income. Thus, we could anticipate that education has something to do with travel motives, but is not necessarily the most important one of all. This is why also the actions taken while on holiday are examined, as well as other factors, such as the destination of the holiday. 10.5.1 Consumption While on Holiday Only a fraction of the activities mentioned in the beginning of this paper, and fully in the appendix, can be seen interrelated with cultural capital and its formation. The appreciation of classical (or other than popular) music and arts could be seen reflecting off the purchasing of music and visiting the museums while on holiday. Buying books shows interest on the literature, though we are not able to specify the type of the literature in question with the help of data. The books purchased can deal either with Donald Duck or ancient culture of the local natives. This is why it is important to take the aspects of purchasing books or music, or visiting museums into analysis. Comparing them with demographics such 190 as education, age and in parallel to various consumption acts, we could possibly decipher the amount of prevailing cultural capital in them. Other possible reflectors of cultural capital are places or events visited while on holiday. Some of the actions mentioned earlier can also be considered less highbrow and less cultural in the sense of cultural capital, and more common to routine tourism activity (such as purchasing local crafts i.e. souvenirs, and which were not taken into account in the analysis). First we take a look into consumption of cultural products while on holiday in general. In Table 1 we can see how much the culturally involved objects are consumed in general among our target groups. After that, in Figure 3, same consumption objects are presented according to region division, the same vein as was done earlier. Table 1. Consumption of culturally involved objects during holiday trip. Percentages, young adults 18–34 yrs (n = 28 701 435). Object of Consumption 18–24 years 25–34 years Total Books 13,9 12,2 12,9 Music 19,6 13,2 15,7 Museums 44,5 47,1 46,1 Churches 31,8 33,8 33,0 Cultural Sites 56,5 62,9 60,4 Concerts 60,3 45,8 51,5 N=28 701 435 According to the Table 1 it seems that some forms of culturally involved consumption are less ordinary than others. Museums and cultural sights can be regarded as the core element of a leisure trip in most cases, which is also visible in the above table. On the other hand it is rather interesting how popular going to concerts seems to be among young adults. Books and music are far rarely consumed while on holiday, and this fact is discussed later. Some distinction between the age groups can be seen for example in music consumption as well as going to concerts. 191 Cultural products 10 19 34 5051 57 11 13 58 15 19 64 5652 60 54 36 1312 36 20 36 35 45 0 10 20 30 40 50 60 70 Books Music Museums Churches Cultural Sites Concerts % Nordic Mediterranean UK and Ireland Central European Fig. 3. Consumption of cultural products while on holiday divided by regions. Percentages, young adults 18–34 yrs (n = 28 701 435). Figure 3 shows us that books and music are not the most popular ways of spending money while on holiday. Even though they are not in the same league as visiting more tourism-related sites and places there can be seen a slight tendency or profile between the regional consumption patterns. However, Figure 3 tells us that the Nordic young adults seem to be quite universally interested in consuming various sites even though they are not top spenders in any other case but visiting museums, exhibitions and archeological sites (referred simply as ‘museums’). Central Europeans are very interested in visiting towns or areas of historical, cultural, or architectural interest, but this is the most common consumption ‘product’ for the other groups as well. Just like when traveling in general was examined, we need to combine educational aspect with tourism consumption. This is done in Figure 4. Figure 4 illustrates the relation between chosen consumption products and the level of education. The level of education seems to be much more distinctive independent variable than the regional variable that was used in previous analysis. Also age, in this case younger age, seems to explain some of the variation (on the basis of cross-tabulation). By this it is meant, that the more time individual has spent time in school, the more likely s/he is to produce the products that have been chosen to be indicators of cultural capital, as far as younger adults are concerned. 192 Books 8 16 19 14 12 8 16 23 12 10 0 5 10 15 20 25 Up to 15 years 16 - 19 years 20 + years Still studying Total 18-25 26-34 Concerts 46 57 74 70 63 43 40 50 68 45 0 10 20 30 40 50 60 70 80 Up to 15 years 16 - 19 years 20 + years Still studying Total 18-25 26-34 Cultural areas 64 49 61 62 57 54 52 59 67 59 0 10 20 30 40 50 60 70 Up to 15 years 16 - 19 years 20 + years Still studying Total 18-25 26-34 Museums 42 50 64 36 50 44 38 51 52 46 0 10 20 30 40 50 60 70 Up to 15 years 16 - 19 years 20 + years Still studying Total 18-25 26-34 Fig. 4. Consumption of cultural products while on holiday according to level of education. Percentages, young adults 18–34 yrs (n = 28 701 435). (NB. Figures are not totally commensurably due to different scaling and proportion.) 193 Cultural areas, this means ‘towns or areas of historical, cultural, or architectural interest’, present on the contrary same phenomenon when it comes to examining senior young adult’s group. Concerts, which could be seen to embody the cultural capital best, are visited nearly in (positive) linear correlation with the length of education. This would suggest, that cultural capital does exist in some role in tourism. 10.5.2 Sand or Sidewalks? Choosing between Sea and City Destination The aspect of holiday destination was also taken into account. As mentioned above, being a tourist in a city rather than on the beach by the sea could be seen as more culturally involved. Thus we could argue, that people favoring city destinations in their traveling are most likely to be interested in the cultural spirit and ethos of that particular place. This could hence mean that cultural capital becomes apparent in city tourism. First, let’s take a look at the trips destined both to the sea and to the city. As can be seen in Figure 5 the sea seems to be far greater a pull factor than the city. This supports the common notion that even today most so called package tours taken are still destined to, sunny and warm locales where there is usually a beach. And where there is a beach, most likely there is also a sea. Education and destination (18-34) 54 65 56 58 61 19 17 26 25 21 51 63 63 54 35 0 20 40 60 80 Up to 15 years 16 - 19 years 20 + years Still studying Total % To the sea To a city Trips in general Fig. 5. The destination of young adults’ holiday, according to age of finishing full- time education, city vs. sea, percentages. (n all travelers = 28 701 435). For the reference the share of the people of having at least one trip (n all young = 50 212 085) 194 Figure 5 thus does not reveal us anything quite interesting, despite the fact that people finishing their full-time education in the age of 16–19 do travel most to sea destinations. But when it comes to the region-based comparison between the two destinations, there seems to be more variation. Figure 6 illustrates the preferences of the whole group of young adults (18–34 years of age) divided by the region of residence. City or sea? (18-34) 61 64 62 61 35 22 21 19 21 43 0 20 40 60 80 Nordi c Medit errane an UK+Ir e Centr al E Total % To the sea To a city Fig. 6. The destination of young adults’ holiday, according to region, city vs. sea comparison, percentages. (n = 28 701 435) In Figure 6 we can see, that sea destination still hangs on to its top position, but the competition becomes quite fierce in the Nordic region. In the other groups the difference is more visible, and also the dominance of the beach resorts thus greater. Let us then examine the city destination solely, since it should bear, according to our assumption, more interesting features in comparison to the sea, and also in the light of cultural capital. Before that it must be reminded that only the first trip (in chronological order) was included in the analysis. The great majority of travelers took only one trip during the year, and those who took trip to both city and sea destination at the same time were scarce. Thus there are some individuals included who have also visited a sea destination but they are taken into account in city comparison as well. First, we would like to know, whether age matters in the decision of taking a city trip. On the daily basis knowledge we could decipher that the older one is, the probable it is for him/her also to have a family. City destinations are not commonly considered to be the most proper destination for family vacations, so we could assume that there are a greater amount of younger travelers in the cities. 195 To a city 39 22 26 24 24 33 22 18 14 19 0 10 20 30 40 Nordi c Medit errane an UK+Ir e Centr al E Total % 18-25 26-34 Fig. 7. City destination and the comparisons between age group and region, percentages (n = 28 701 435) As was assumed, and in Figure 7 can be seen, the tendency for younger adults to travel to city is greater than for the ‘senior’ young adults. This division does not still reveal if cultural capital has anything to do with this tendency. In order to examine also that, education variable needs also to be taken into account. In Figure 8 the region-based subgroups were studied separately in the light of the education level. This enables us to clarify the possible reasons why disparities between subgroups occur. City 8 22 16 18 33 18 17 15 37 26 38 22 38 25 40 22 0 10 20 30 40 50 Nordic Mediterranean UK+Ire Central E % Up to 15 yrs 16 – 19 yrs 20 + yrs Still studying Fig. 8. City destination and the comparisons between regions and education level, percentages (n = 28 701 435) 196 The amount of educations seems to have on some occasion significant role when choosing city as a holiday destination. Most visible disparities emerge both in Nordic regions and in the UK and Ireland. 10.6 Conclusions Now we have shed some light on the cultural capital and it’s occurrence and possible use in interpreting cultural tourism, as well as described the travel patterns of young Europeans, taking the Nordic young adults as a reference group in most occasions. As the analysis proceeded some interpretations were presented, but now we would like to conclude by reviewing the results of this study. 10.6.1 Peculiar young travelers from the Nordic countries It seems according to cross-tabulation that travel patterns of the young Scandinavians differ from those from other part of Europe. As we have seen in examining the length of education in regional light, the Nordic youth both spends the longest time in educational institutions and also, in addition to that, takes most often trips. Purchasing books was not as popular among Nordic young adults as it was among the other young Europeans. One reason for this could be language, but it would not still explain very well the less interested attitude towards these products by the Scandinavians. Books and music are in general still rather little consumed (or: purchased) while on holiday; even the most eager readers – according to this data – from central Europe do not quite reach up to 15–18 per cent. When it comes to visiting museums, exhibitions and archeological sites the Nordic youth shows most interest compared to the others. Places of worship are quite even in popularity among the whole target group, even though one could predict that rather religious culture and way of life in the Mediterranean region have its impact on this matter. Museums and churches are more eagerly consumed, most likely because they are seen more commonly to be routine consumption targets while on holiday in general. Still the variation between the places and events visited is clearly profiled when it comes to comparing the young Europeans by regions. It seems according to our results, that the Nordic youth travel most eagerly, when compared with their counterparts from other regions of EU. Perhaps mostly due to economic factors (the older the bigger incomes) the group of young adults of 26 to 34 years of age tends to travel more than the even 197 younger adults. Despite of this fact, even the younger adults from Scandinavia travel more often than the older young people from other regions. This could either be seen as a reflection of higher educational and thus cultural capital, which can be derived from higher age of finishing full-time education in the north. This aspect would then suggest that the more cultural capital one possess, the more eager s/he is to travel and thus express it via tourism. On the other hand this could simply refer to different kind of attitude towards traveling which prevails in the north due to, for instance, geographical factors. A third explanation for more intense travel frequency of Nordic young adults can be the fact, that also trips to summer cottage was considered as holiday trips. There is an extra notification about this in the Finnish questionnaire, whereas in English counterpart no statement of that kind exists. This of course reflects the unique feature of Finns and their summer cottage “culture”. This analysis shows that there are numerous possibilities for further study in the field of culturally involved consumption while on holiday. One of the most intriguing approaches would be examining the very essence of city tourism. Who is a city tourist? What does s/he consume there? Why people from some region prefer city destination more than some others? Does education increase the inclination to start liking city traveling more than anything else? Various answers remain still unanswered, as there are so many aspects that need to be taken into consideration. This paper has nevertheless accomplished its aims by suspecting, that cultural capital plays some role in – at least young European adults’ – travel patterns and tourism. Via examining such consumption products as going to concerts that could be seen exhibiting cultural capital we have both acknowledged cultural capital’s prevalence, and we have also shed light on some of the today’s travel patterns of European young adults. 198 References Alasuutari, Pertti (1997) Kulttuuripääoma summamuuttujan valossa (Sum Variable in Light of Cultural Capital). Sosiologia Vol. 34, Nr.1, 3–14. Bocock, Robert (1995) Consumption. Routledge: London. Bourdieu, Pierre (2002) The Forms of Capital. In: Readings in Economic Sociology, ed. by Nicole Woolsey Biggart, 280–291. Blackwell: Oxford. [Original publication (1986): In: Handbook of Theory and Research for the Sociology of Education, ed. by John G. Richardson. Greenwood Press: Westport, Conn.] Bourdieu, Pierre (1994) Järjen käytännöllisyys. Toiminnan teorian lähtökohtia. Vastapaino: Tampere. Bourdieu, Pierre (1984) Distinction. A Social Critique of the Judgement of Taste. Routledge: London. Härkönen, Juho (2002) Kotitaloudet, työmarkkinat ja köyhyys – Kotitalouksien työmarkkinaosallistuminen ja köyhyysriskit Euroopan unionissa. (Households, Labour Markets and Poverty.) In: Tieteellisen sosiologian puolesta. Taloussosiologian tutkimuspäivä 23.11.2001, ed. by Timo Toivonen, 31–65. Turun kauppakorkeakoulun julkaisuja, sarja Keskustelua ja raportteja, 3:2002. Turun kauppakorkeakoulu: Turku. Katz-Gerro, Tally – Shavit, Yossi (1998) The Stratification of Leisure and Taste: Classes and Lifestyles in Israel. European Sociological Review, Vol 14, No. 4, 369–386. Kyrö, Matti (2002) Finnish abstract of the OECD Report “Education at a Glance 2002”. Ministry of Education: Helsinki. Meetham, Kevin (2001) Tourism in Global Society – Place, Culture, Consumption. Palgrave: New York. Melich, A. (1999) Eurobarometer 48,0: Holiday travel, October-November 1997. [Electronic data].Brussels: INRA (Europe) [producer], Ann Arbor, MI: Inter-university Consortium for Political and Social Research; Köln: Zentralarchiv für Empirische Sozialforschung; Yhteiskuntatieteellinen tietoarkisto [distributors]: Tampere. Mäkelä, Johanna (1994) Pierre Bourdieu – erottautumisen teoreetikko. (Pierre Bourdieu – The Theoretician of Distinction) In: Sosiologisen teorian nykysuuntauksia, ed. by Risto Heiskala. Gaudeamus: Tampere. 199 Pronovost, Gilles (1989) The Diversity of Social Time: the Role of Institutions. Current Sociology, Vol. 37, No. 3, 37–62. Räsänen, Pekka (2002) Imaginäärinen kulutus, rakenteelliset olosuhteet ja luokkaidentifikaatio. (Imaginary Consumption, Structural Circumstances and Class Identification.) In: Tieteellisen sosiologian puolesta. Taloussosiologian tutkimuspäivä 23.11.2001, ed. by Timo Toivonen, 79–99. Turun kauppakorkeakoulun julkaisuja, sarja Keskustelua ja raportteja, 3:2002. Turun kauppakorkeakoulu: Turku. Richards, Greg (1996) Introduction: Culture and Tourism in Europe. In: Cultural Tourism in Europe, ed. by Greg Richards, 3–17. CAB International: Oxon. Ritzer, George (2001) Explorations in the Sociology of Consumption. Fast Food, Credit Cards and Casinos. SAGE: London. Sanaksenaho, Sanna (2002) Kulttuurinen ja sosiaalinen pääoma Suomessa: Ryhmittelyanalyyttinen näkökulma. (Cultural and Social Capital in Finland from the Cluster Analysis’ Point of View.) In: Lama ja luottamus, ed. by Kaj Ilmonen, Anne Kovalainen and Matti Siisiäinen. Forskiningrapporter från Svenska handelshögsskolan 55. Hanken: Helsinki. Selänniemi, Tom (1997) Suomalaiset etelässä. (The Finns in the South) In: Maailmalle, maailmalta. Kulttuurisia katsauksia, ed. by A. Syrjänen and I. Westermarck. WSOY: Juva. Toivonen, Timo (2001) Are the Differences in Holiday Frequency between Countries Only Structural? A Comparison of 15 EU Countries. In: You’ll Never Walk Alone. Studies on International and National Developments in Tourism, 39–59. Turun kauppakorkeakoulun julkaisuja, sarja Keskusteluja ja raportteja 4:2001. Turun kauppakorkeakoulu: Turku. Toivonen, Timo (2000) Tulojen ja kohortin vaikutus turismikulutukseen laman aikana. (The Impact of Income and Cohort on Tourism Consumption during the Economic Depression.) Sosiologia, Vol. 37, No. 1, 48–59. Toivonen, Timo (1999) Empiirinen sosiaalitutkimus. Filosofia ja metodologia. (The Empirical Social Research. The Philosophy and Methodology.) WSOY: Helsinki. Urry, John (1996) Tourism, Culture and Social Inequality. In: The Sociology of Tourism. Theoretical and Empirical Investigations, ed. by Y. Apostolopoulos et. al, 115–133. Routledge: London. Wilska, Terhi-Anna (1999) Survival with Dignity? The consumption of Young Adults during Economic Depression: A Comparative Study of 200 Finland and Britain, 1990–1994. Turun kauppakorkeakoulun julkaisuja, sarja A-3:1999. Turun kauppakorkeakoulu: Turku. WTO Official Website. Methodological notes on concepts related to tourism. , 10.4.2003. 201 Appendix Key questions and variables in data Q64: On this first holiday (holiday of 4 nights or more), where did you (or will you) go to…(several answers possible= SAP)? To the sea To the mountains To the countryside To a city Q71C: What types of local products do you ever buy when on holiday (SAP)? Local craft products Clothes Books Music Food products Q72C: What kind of places do you ever visit when on holiday (SAP)? Museums, exhibitions, archeological sites National parks, nature reserves Theme parks Spas, health centres Churches, temples, mosques, synagogues, etc. Towns or areas of historical, cultural, or architectural interest Industrial sites / factories Q73: On holidays, do you go to…(SAP)? a) religious events/ services b) sports events c) concerts/music events d) traditional local events e) other local events D8. How old were you when you finished your full-time education? Up to 15 years 16 – 19 years 20 + years Still studying 202 Kirjoittajien yhteystiedot – The Contact Information of the Authors Valtiot. yo. Antti Gronow Opiskelija/ Student Helsingin yliopisto/ University of Helsinki Sosiologia/ Sociology antti.gronow@helsinki.fi VTM Jani Erola Tutkija/ Researcher Turun kauppakorkeakoulu / Turku School of Economics and Business Administration Taloussosiologia/ Economic Sociology jani.erola@tukkk.fi KTM, FM Leena Haanpää Assistentti/ Research Associate Turun kauppakorkeakoulu / Turku School of Economics and Business Administration Taloussosiologia/ Economic Sociology leena.haanpaa@tukkk.fi VTM Antti Honkanen Yliopettaja/ Principal Lecturer Vaasan ammattikorkeakoulu/Vaasa Polytechnic Matkailu- ja ravintolapalvelut/ Tourism and restaurant services antti.honkanen@puv.fi VTM Juho Härkönen Tutkija/ Researcher European University Institute Department of Political and Social Sciences Juho.Harkonen@IUE.it VTM Pekka Mustonen Tutkija/ Researcher Turun kauppakorkeakoulu/ Turku School of Economics and Business Administration Taloussosiologia/ Economic Sociology pekka.mustonen@tukkk.fi M.A. Sueila Pedrozo Tutkija/ Researcher National Society for Teaching and Research in Management Rio de Janeiro, Brazil sueila@pedrozo.com VTT Pekka Räsänen Tutkija/ Researcher Turun yliopisto/ University of Turku Sosiologia/ Sociology pekras@utu.fi VTT Timo Toivonen Professori/ Professor Turun kauppakorkeakoulu/ Turku School of Economics and Business Administration Taloussosiologia/ Economic Sociology timo.toivonen@tukkk.fi KTM Taru Virtanen Tutkija/ Researcher Turun kauppakorkeakoulu/ Turku School of Economics and Business Administration Taloussosiologia/ Economic Sociology taru.virtanen@tukkk.fi 203