Scholarly article on topic 'Economic, Cultural and Social Factors Influencing the Development of Gay Businesses and Places: Evidence from the European Union / Ekonomické, Kulturní A Sociální Faktory V Pozadí Rozvoje Gay Podniků A Míst: Situace V Evropské Unii'

Economic, Cultural and Social Factors Influencing the Development of Gay Businesses and Places: Evidence from the European Union / Ekonomické, Kulturní A Sociální Faktory V Pozadí Rozvoje Gay Podniků A Míst: Situace V Evropské Unii Academic research paper on "Social and economic geography"

CC BY-NC-ND
0
0
Share paper
Academic journal
Moravian Geographical Reports
OECD Field of science
Keywords
{""}

Academic research paper on topic "Economic, Cultural and Social Factors Influencing the Development of Gay Businesses and Places: Evidence from the European Union / Ekonomické, Kulturní A Sociální Faktory V Pozadí Rozvoje Gay Podniků A Míst: Situace V Evropské Unii"

MORAVIAN GEOGRAPHICAL REPORTS

DOI: 10.2478/mgr-2014-0014

3/2014, Vol. 22

ECONOMIC, CULTURAL AND SOCIAL FACTORS INFLUENCING THE DEVELOPMENT OF GAY BUSINESSES AND PLACES: EVIDENCE FROM THE EUROPEAN UNION

Koessan GABIAM, Michal PITONAK

Abstract

The late 20h century emergence of gay neighbourhoods and their related businesses has been examined by a number of researchers, but few have tackled this issue on a larger scale compared to case studies in a national context. This study considers the development of European gay businesses and gay non-residential places, using principal components analysis on data from the Spartacus International Gay Guide of 2007. This is an unusual quantitative approach in the largely qualitatively-dominated field of geographies of sexualities. It has identified gay commodification and gay (in)visibilities as the most likely factors of spatial diversity in observed gay places. These two identified dimensions are then analysed in terms of their linkages, specificities and regional importance. Subsequently, the interactions between economic, cultural and social factors at stake in the development of gay business and non-residential places across Europe, are evaluated.

Shrnuti

Ekonomicke, kulturni a socidlnf faktory v pozadirozvoje gay podniku a mist: situace v Evropske unii

Vznik gay ctvrti a s nimi spojenych gay podniku na konci 20. stoleti byl objektem zajmu mnoha vyzkumniku, ale pouze malo z nich - ve srovnani s pripadovymi studiemi v kontextu narodnim - resilo tuto otazku na urovni celoevropske. Tato studie se zabyva rozvojem gay podniku a ne-rezidencnich mist za pouziti analyzy hlavnich komponent aplikovane na datech ziskanych ze Spartacus International Gay Guide 2007. Tento kvantitativni pristup je v oboru prevazne kvalitativnich geografii sexualit ponekud nezvykly. Diky tomu byly identifikovany gay komodifikace a gay (ne)viditelnosti jako nejpravdepodobnejsi faktory vysvetlujici prostorovou diverzitu zkoumanych gay mist. Tyto dva faktory jsou analyzovany z hlediska jejich spojitosti, specifik a regionalni vyznamnosti. Nasledne jsou hodnoceny interakce v ramci ekonomickych, kulturnich a socialnich faktoru pusobicich pri rozvoji gay podniku a gay ne-rezidencnich mist napric Evropou.

Keywords: gay businesses and places, commodification, gay (in)visibilities, European Union

1. Introduction

Geography has evolved greatly and now spans various cultural, economic, political and societal domains, where its true potential can be recognized specifically in its ability to understand connections between these domains. As with the development of human societies, the character of geography has changed and is influenced by more and more complex interrelations. This paper is grounded in the geographies of sexualities, a geographical sub-field which developed originally in British and North American geographical academia (Brown, Knopp, 2002; Hubbard, 2001; Rose, 1995) under the larger influences of post-positivism and the 'new' cultural turn in geography from the 1970s. Later this field expanded to other countries in Western, Mediterranean and Central Europe, Australasia and Latin America. Geographies of sexualities are nowadays diverse fields of study covering issues ranging from homosexuality (Duncan, 1996; Binnie, Valentine, 1999), heterosexuality (Oswin, 2008), bisexuality (Hemmings, 2002) and transgenderism (Doan, 2007; Browne, Nash, Hines, 2010; Nash, 2010). We cannot provide a complete summary of its genealogy in the present article; however, a short introduction to the associated main concepts and cornerstones is relevant here.

The first studies associated with geographies of sexualities were set in North American urban sociology. Societal conditions after the turbulent 1960s allowed a focus on

visible and material sexual urban contexts, specifically the political-economic development of so-called "gay ghettoes". These visible concentrations of both gay commercial spaces and gay residents were identified in major American cities such as New York, Chicago and San Francisco (Levine, 1979; Castells, 1983; Knopp, 1987). These studies were very descriptive, while focusing on processes such as urban renewal or gentrification and the role of gay populations and gay ghettoes in them. Any focus on lesbian geographies remained silent as their spatial dimension required more sensitive approaches to develop (Rothenberg, 1995; Johnston, Valentine, 1995). These first studies did not introduce any critical analyses beyond political-economic debates, but their main merit for later studies was that they brought the topic of sexuality (homosexuality) into the focus of academia and helped it become a new proper subject matter.

In the subsequent years in the late 1980s and 1990s, important developments of critical social theories, mostly feminist and queer theories but also the "general" postmodern and poststructuralist ways of thought, provided important background for more rigorous geographical research (Bell, Valentine, 1995; Duncan, 1996).

Power relations between people of normative (hetero) and non-normative (LGBTIQ...) sexualities started to be problematized. Previously unproblematic, heterosexual

practices such as holding hands, kissing or even living together had been identified as naturalized and inherently sexual (Valentine, 1993). In contrast, these same practices in same-sex couples were automatically perceived as sexual and "peculiar". Heteronormativity, as a normative regime, was therefore suggested as a structure which was about to be academically recognized. Similarly, as feminist scholars identified normative patriarchy, sexuality and queer scholars identified heteronormativity (Pratt, 2009). Space itself was recognized as vastly sexualized and normalized under the rule of heteronormativity, where sexual-others could not safely exist and manifold exclusions were achieved by creative regulative regimes (Howell, 2009: 122). Understandably, many geographers focused on the ways and practices of social (re)production of heteronormativity, and specifically on the ways in which space was being (re)produced as (hetero) sexual or heteronormative (Browne, Lim, Brown, 2007:2; Gorman-Murray, Waitt, Johnston, 2008:236).

The most influential factors were the theoretical works of post-feminist literary philosopher Judith Butler and (post)structuralist philosopher Michel Foucault, which informed new associated geographical scholarship (Foucault, 1995, 1980; Butler, 1990, 1993). Power as a factor was no longer "out of the theoretical reach", and the duality of 'structure and agency' continued to be studied in dialectical ways (Butler), human subjectivity and identity were deconstructed, ways beyond normative thinking and the perspectives of (sexual) majorities started to be sought. The loose association of approaches trying to reach these goals may be called queer theory1.

During the 1990s most geographers still retained their urban focus, specifically in Europe where inner districts of major cities started to show interesting developments, or even explosions, of mostly gay commercial spaces. These urban developments, the so-called "gay villages", were closely connected to political, societal and economic changes in the late 1980s and 1990s. Geographers have begun deciphering various factors which have enabled the visible existence of gay spaces. The former term "gay ghettoes" became obsolete, pejorative and imprecise, and not only in the European context. Economic and political factors have been analysed in detail, especially those connected with the 'pink economy', which is a complicated term that might refer to a capitalist utilization of a disadvantaged groups' position (here non-heterosexual people). In other words, the pink economy might refer to a market niche available in heteronormative culture (Binnie, Skeggs, 2004), where certain environments (cities), with certain political-economies (democracy, capitalism), enabled the formation of gay- or lesbian-oriented commercial venues. Contemporary research in this field suggests (Bassi, 2006) that we should not jump to premature conclusions about the pink economy, as there might be other cultural-economic factors involved. We believe that our paper sheds some light on this issue as well.

What seems obvious, however, is that the developments of gay commercially-oriented venues did not cater to all non-heterosexuals, but indeed produced certain sexual-cultures (mainly gays') with their specific identity and inherently exclusionary politics. Thus the concept of homonormativity

suggested itself. Certain bodies such as male, white, young, handsome and willing to consume, were preferred to socialize in new glamorous gay bars, cafés or restaurants, while other bodies were implicitly or explicitly excluded (Gorman-Murray, Waitt, Johnston, 2008). As Lisa Duggan had put it: "homonormativity... does not challenge heterosexist institutions and values, but rather upholds, sustains, and seeks inclusion within them" (Duggan, 2003).

Many spaces such as home (Johnston, Valentine, 1995), gay and lesbian bars (Valentine, Skelton, 2003) were studied in relation to space and human identity formation, but the general economic, cultural and social factors influencing the development of such places remains largely understudied.

It is relatively self-evident now that the previously dominant spatial science, with its rigid paradigms and conceptualizations of human behaviour, cannot provide explanations to these socio-spatial developments. Space, place and landscape are no longer perceived as Cartesian or deterministic, but rather as socially produced and shaped by complex power relations. The theories indicated above have re-conceptualized the role of human beings and their agency, and wider understandings of the larger structural influences imposed on them. It was mostly feminist and queer theories in geography which raised our understanding that human beings are not all alike but different and socio-spatially structured in complicated ways, on the basis of their gender, sex, race, sexuality and the other axes of human difference (Valentine, 2007). The socially constructed nature of structures such as patriarchy and heteronormativity seem to have been already vindicated (Hubbard, 2001; Kirby and Hay, 1997).

In the 2000s, several geographers underlined important methodological issues in mapping quantitative data of gay and lesbian lives and migration, by critically using national censuses such as in the United States (Brown, Knopp, 2006; Cooke, Rapino, 2007) or in Australia (Gorman-Murray et al., 2009). Those studies gave birth to a rich reflection on the intersection of queer theory and social science research (Browne, 2010). Furthermore, as emphasized by Michael Brown and Larry Knopp: "serious engagements with sexuality necessitate a careful reconsideration of some fundamental ontological, epistemological and methodological issues" (2003:313). This paper raises similar issues, bringing methodological questions into the dialogue with quantitative studies and the qualitative approaches, interrogating how global and national forces2 interact with a marginal and cultural phenomenon such as the development of gay businesses and places, in terms of their structurally different distribution throughout the EU, Norway and Switzerland3. The methodological approaches in sexuality and queer geography scholarship may be diverse, but they are largely qualitative. We believe that this domination may be contested as unnecessary.

We now proceed with a methodological section where we describe our primary source of data, together with the chosen quantitative analytical tool - principal components analysis (PCA). Then we focus on the selected variables and discuss some of the initial observations. From these initial results, we identify two main principal components (CP1 and

1 For more about the queer theory, see Jagose, 1996.

2 Being aware of other involved forces on different levels, such as local and city authorities, etc.

3 Switzerland and Norway are not members of the EU, but their regional-political orientation is European. Therefore they had been considered in this research as if they had the same status as other EU states.

CP2) and focus on their interpretation and on an in-depth description of the results. We then conclude that issues of commodification and (in)visibilities are of major influence in structuring the spatial diversity of gay businesses and places. It is then the search for factors of gay (in)visibilities, commodification and gay place/spatial diversification, that lays behind our research interest.

2. Methodology

This paper focuses on the relative importance of gay facilities by analysing the proportion of the different types of 'gay places' in European cities, using quantitative methodology. The term 'gay' is used here for self-defined male homosexuals, whereas self-defined homosexual females are referred to as lesbians. Used as an adjective for the discussed places, it refers to those that permit self-defined homosexuals to live out their sexuality openly. From this perspective, patronising such places helps build a gay identity and culture, one in which homosexuality may be fully performed by discourses or acts.

For data analysis we utilized a principal components analysis (PCA), a mathematical version of 'factor analysis' which transforms a number of correlated variables, here indexed by the number of various 'gay places', into a (smaller) number of uncorrelated variables called principal components (see section 3). The term 'gay places' is used throughout this article for 'various gay places registered in the Spartacus International Gay Guide in its 2007 edition'. We utilized this 'guide' as a database for secondary data. This database has become a useful basis for other geographers and social scientists worldwide, so we are convinced that it is useful for analysing the spatial distribution of gay facilities through the end of the 20th and the early 21st century (Blidon, 2007; Salinas, 2007; Leroy, 2005).

Despite some criticism for often being slow in rectifying out-of-date address details, the Spartacus database was utilised for its uniqueness in presenting a global list of gay businesses, associations and cruising locations4 alike. Therefore, we assumed that the selection criteria for listing were standardised, and hence comparable for all European countries. In this way the probability of errors was similar in all studied regions of the EU, Norway and Switzerland: for example, there was the same probability for listing a closed gay business or not including an active one in the study region.

Further, it is important to state that the Spartacus Gay Guide does not specifically cater to lesbians and the other sexual or gender minorities (bisexual, intersexual, MSM5, transgender, etc.) and only in its first edition in 1974 did the authors include a focus on lesbian venues. Due to the Spartacus Gay Guide's narrow 'gay' focus, we have to limit the validity of most of our conclusions to 'mainstream gays'. At the same time, however, we bear in mind that people with other sexual identities are potential customers

of the studied venues, and that most of these neither exclude them nor should they. It has been recognized in various studies that the potentials for sustaining solely lesbian venues are low, due to lesbians' weaker economic power (Adler, Brenner, 1992; Bondi, 1998; Duncan, 1996; Valentine, 1993; Weightman, 1981; Moran, 2001: 410). It is probably because of this situation that we did not find any comparable database for lesbian venues which would complement the data for 'gay venues' listed in Spartacus Gay Guide. Admittedly then, parts of lesbian geographies remain largely undiscovered by geographers and deserve further investigation. Significant contributions to lesbian research have been made during the last decades by several geographers such as Gill Valentine (1993a) or Nadine Cattan and Anne Clerval (2011).

While being aware of certain limitations, we are convinced that this methodology has enabled us to perceive overall patterns that would stay hidden or implicit if a qualitative methodology was used. The chosen methodology sheds light on some underlying factors of the emergence and development of 'gay places' in large cities, while staying at a European scale. This offers an opportunity to test some hypotheses proposed by the large body of scientific research on gay, lesbian and queer geography during the last decades. Such research suggests some explanations for the development of gay residential and commercial neighbourhoods - by gentrification and central urban regeneration processes in England (Collins, 2004); for Spain, Garcia Escalona (2000) uses the term 'ghetto' when dealing with Madrid's Chueca's case; and in France the 'Marais' case in Paris is explored by Michael Sibalis (2004). Although most of the existing research makes use of case studies and local or national data sources, this article utilises a rich international database for the purpose of uncovering and analysing complex mechanisms and trends which may lead to the diverse and uneven development of gay spaces in Europe.

3. Data analysis

The focus for the present study was the proportion of types of gay facilities in European cities with at least 100,000 inhabitants. Each city having at least 100,000 inhabitants presented in the Spartacus Gay Guide 20076, even if there was just one 'gay place', was integrated into the database for the PCA. Both criteria led to a total of 377 cities.

Preliminary analyses on these sampled cities did not lead to a clear identification of possible explanatory factors. Therefore, our subsequent analysis focused on 168 European cities, which are comprised of urban areas with at least 100,000 residents (defined in the Morphological Urban Area as fixed by IGEAT-ULB in the ESPON study on urban functions)7, and where at least 10 'gay places' were located. The latter threshold is set in order to meet 'normal' analytical requirements for a database in terms of statistical

4 Cruising places are exterior spaces (e.g. parks, wooded areas or service areas) where men can meet other men for brief sexual encounters.

5 Men who have sex with men.

6 We have counted up the number of 'gay places' listed in the Spartacus Gay Guide issued in 2007 and established a Europe-wide database. We were not given permission by the Spartacus Gay Guide editor to publish such a database nor set it up as an open source resource, however.

7 ESPON is the European Spatial Planning Observation Network, which researched the delimitation of urban areas in Europe. Since the definition given to a city or to an urban area differs from country to country, one aim of ESPON, among others, is the establishment of the number of inhabitants in urban areas on the basis of multiple data sources whilst utilizing a common methodology in the data collection across all European countries.

power. Interestingly, several authors have posed the question of a threshold value for the sustainability of gay business or to represent its significance in cities. It has been suggested that at least 50,000 people were needed as a source population for sustaining a gay venue, such as a gay bar (Miller, 2009). Having more than 20 'commercial establishments catering to gays' seemed to Dennis Altman (1996) an acceptable criterion to select cities worldwide, where gay commercial expansion was occurring. Such thresholds should be fixed by taking the scale of the analysis into account.

We based our PCA analysis on 27 different categories of gay places, representing businesses and locations from the Spartacus Guide in its 2007 edition. Some of the variables have been aggregated, whereas other guide categories have been split for reasons of geographic location. This

study focuses on urban areas, therefore the nation-wide categories of the Spartacus Guide (like national gay info, publications, companies or help lines) were not used. In addition, any on-line categories without a street-level address were not included. Regarding the cruising (see footnote 5), two categories were set up by distinguishing between cruising places located in swimming pools, added to the beach category, and all other cruising places. The categories of culture, museum, archives and monument were merged, because of their limited size and close relationship to culture in general. The averages and standard deviations of the 27 categories are shown in Table 1. The most frequent categories in the 168 studied cities were bars (22% of the gay places)8, cruising places (16%), dance clubs (9%) and saunas or bathhouses (7%).

Category Average presence among 168 cities Standard deviation PC 1 loadings PC 2 loadings Identifying number in Fig. 2

Culture / monument / archives / museums 0.27 0.93 0.20 - 0.31 7

Fashion shops 0.33 0.89 0.53 - 0.08 10

Massage 0.33 0.61 0.16 0.40 19

Fitness studios 0.35 0.83 0.52 - 0.19 11

Drag shows 0.35 0.82 0.36 0.27 25

Travel agencies 0.48 0.94 0.49 0.11 27

Cinemas/Blue movies 0.60 1.86 - 0.13 - 0.49 3

Escorts/Studios 0.61 1.28 0.18 - 0.19 9

Local publications 0.73 1.17 0.51 0.22 18

Private accommodation 0.88 1.41 0.45 - 0.16 21

Leather/Fetish shops 0.90 1.55 0.44 0.08 17

Bookshops 1.04 1.56 0.29 0.03 4

Health groups 1.08 1.85 - 0.03 - 0.44 15

Gay tourist info. (not Internet) 1.81 2.44 0.05 - 0.56 12

Apartments 1.86 2.81 0.54 - 0.21 1

General groups / Other groups 1.90 2.93 - 0.12 - 0.27 13

Guesthouse 3.19 4.78 0.21 - 0.42 14

Men's clubs 3.44 3.16 0.47 0.14 20

Cafes 4.00 3.94 0.10 0.30 5

Cruising places (beaches and swimming pools) 4.65 6.05 - 0.33 - 0.38 26

Restaurants 5.30 4.89 0.58 - 0.14 22

Sex shops and Blue Movies 6.01 4.74 - 0.21 0.17 24

Hotels 6.09 6.01 0.57 - 0.15 16

Saunas/bathhouses 6.63 4.66 - 0.24 0.13 23

Dance clubs 9.43 6.01 - 0.25 0.11 8

Cruising places 15.56 13.14 - 0.68 - 0.10 6

Bars 21.67 10.86 0.07 0.64 2

Hypothesis category 1: Total gay places Not in PCA Not in PCA A

Hypothesis category 2: Urban area population Not in PCA Not in PCA B

Tab. 1: Selected aggregate variables (categories) included in the PCA

8 Concretely, every city of the sample (having more than 100,000 and having at least 10 places), has at least two bars, generally speaking. Further, there are 10 cities from the sample that have no bars among their gay facilities.

When assessing the geographical distribution of 'gay places' at the European macro-regional level, it became clear that there is a relationship between population size and the number of registered 'gay places'. Table 2 displays the total number of 'gay places' for selected European countries: note that the more populated European states, such as Germany, Spain, France, Italy or the United Kingdom (UK), also have the highest counts. As will be obvious to many readers here, there are however great discrepancies between a country's population and its number of 'gay places'. For example, consider Poland with 40 million inhabitants and Czechia with 10 million inhabitants - respectively having 92 and 97 gay places. Such discrepancies are about to be explained by factors revealed in this study.

This relationship was considered further by analysing the correlation between urban area population, as given in the ESPON study (see footnote 7), and the number of 'gay places' per urban area.

Country Number of 'gay places' Number of gay places (cities having at least 10 'gay places')

Germany 1245 1138

Spain 995 685

France 863 729

Italy 685 479

UK 555 236

Netherlands 309 265

Belgium 234 174

Switzerland 150 146

Greece 121 39

Austria 106 81

Portugal 105 99

Sweden 101 40

Czechia 97 84

Poland 92 62

Denmark 69 62

Hungary 63 54

Ireland 49 49

Romania 49 19

Finland 48 35

Norway 39 28

Bulgaria 36 21

Slovakia 24 19

Estonia 23 19

Slovenia 21 15

Lithuania 20 10

Cyprus 19 12

Latvia 17 17

Luxemburg 14 14

Malta 7 0

Tab. 2: Number of 'gay places' per selected European country

The correlation is quite strong at 0.66 for the preliminary sample of 377 cities. This explains why the European megalopolis, the so-called blue banana, a large area stretching from Southern England to the large cities in Benelux countries, Western Germany, Switzerland and Northern Italy, is immediately obvious (see Fig. 1). Some important exceptions exist, however. For instance there is a group of large capital cities (specifically in the Central European region) with relatively small numbers of gay facilities. The only exceptions are Budapest and Prague, which are relatively well-served. The opposite situation can be observed at sea resorts with mostly smaller populations, such as Sitges, Ibiza, Benidorm or Torremolinos (Spain), Mykonos (Greece), as well as Blackpool and Brighton in the UK. The situation at the sea resorts can be relatively easily explained due to their commercial-tourist orientation and not to a permanent residential orientation.

Principal components analysis (PCA) was used for the purpose of identifying the most important linkages between the variables. PCA is helpful to calculate, visualise, and comprehend such linkages. This statistical analysis aims to select a subset of variables from a larger set, based on which original variables have the highest correlations with the principal component. Geometrically, the produced component may be understood as the axis or vector passing through the maximum variation (variance) of the projected values of the original observations (data points). Successively, new independent components are created with decreasing variance accounted for in the residual variance matrix. Further developments in PCA are demonstrated by I. T. Jolliffe (2002). The results of the PCA are displayed as new variables (the principal components), which synthesise the included information in the initial 27 variables. Here, it is useful to treat our database because it consists of numerous categories (27) observed in 168 cities. This fact makes the database complex if we want to find out directly whether any geographical opposition exists between cities and/or countries regarding the kind of existing gay facilities they have.

This study examines only the two most important principal components. These two components account for a total of 57% of all variance (i.e. the information in all original 27 variables), with the first component accounting for 37% of the variance, and the second for 20%. Furthermore, the component matrix represents a projection of the two first components' plan, which enables one to visualise the position of each variable when projected on the plan of the two new dimensions or components, CP1 and CP2 (see Fig. 2 and detailed legend from Tab. 1). On the basis of the PCA results, we focus our attention on the highest correlations between variables and components which resulted in the identification of the two first components: a detailed discussion is presented in the following section.

4. Commodifying 'gay places'

The first dimension that arises from the analysis, the first principal component (CP1), is positively correlated with restaurants (Fig. 2, label 22), hotels (label 16), and apartments (label 1). On the opposite side of the axis, cruising places (Fig. 2, label 6) have the highest negative correlation (— 0.68) with CP1. As places of accommodation (comfortable) and cruising places (uncomfortable) obtain the highest correlations, this relation might be described as a gay 'comfort' axis. Since accommodation capacity is linked to tourism, an interpretation of CP1 as a touristic dimension is also feasible. In his study, Howard

Fig. 1: Total number of gay places and businesses in major urban areas in the EU, Norway and Switzerland Sources: Spartacus Gay Guide (2007); made with Philcarto: http:Hphilcarto.free.fr

EU, Norway, Switzerland: gay places and businesses (PCA, saturations)

/ \

/ Bars(; ) \

/ \

/ / N

/ \

♦ 25 18 ■ \

1 24 ' .2 0 \

23 4 ♦ ♦ i

0 -0 .9 -0 8 -0 7 -0 .6 -0 5 -0 .4 -0 3 -0 .2 -0 .1 0 0 0 1 0 2 0 3 0 4 0 5 0 ♦10 6 0 7 0 8 0 9 1

I c ruising )laces(6 ) 9 21 ♦ ♦♦ ♦ Af testaurc artmen nts(22) s(l) /

\ H * ♦Ho els (16 /

\ 2 ♦ /

\ S Blue M ovies(3 ♦ • 15 ♦ 14 /

\ Gay in ♦ o(12) /

\ /

\ /

Fig. 2: Variable loadings on the first two principal components, data on gay places and businesses in EU, Norway and Switzerland. Source: Computed by authors'

Hughes (2006) reported a survey that portrayed hotels as the most used accommodation by gay and lesbian travellers; nevertheless, bisexual, transgendered and the other sexual minorities have access to them as well. Interestingly, hotels have a higher correlation with the first component than apartments or private accommodations, even if they supposedly offer more privacy to gays. Therefore, 'comfort' may be considered as a consequence of a touristic dimension. On the other hand cruising places appear to be more frequent in smaller towns (from the pool of over 100,000 inhabitants cities) than in larger cities. Therefore, we can also think that the PCA opposes small towns to larger cities and that CP1 is also interpretable as an urban hierarchy axis (see Fig. 2, the correlation of the variable 'Total 'gay places''/ short label A).

The variable 'urban area population' (short label B) was also projected on the components graph as it shows a correlation of 0.6 with CP1. This result is weaker in absolute value than the correlation of — 0.68 for the cruising variable, but this can be explained by the unique significance of seaside resorts, which are well-served while being less densely populated. Nevertheless, an interpretation of the first component as a touristic dimension, together with the urban hierarchy, is not necessarily contradictory. Here, it is useful to keep in mind that the correlation coefficient between overnight stays9 (European Cities Tourism, 2011) in cities and the population of the urban area is 0.55, which shows a strong link between the urban hierarchy and urban tourism.

Besides the factors discussed above, commodification processes of gay culture are at stake in the entrepreneurial cities. As mentioned in the introduction, the commodification of gay culture is considered by some authors (Binnie and

Skeggs, 2004; Hughes, 1997) as a tool for attracting global capital flows into cities by funding the rise of gay villages in city centres (e.g. the gay village in Manchester, UK). Therefore, over-equipped cities with gay facilities can be seen as those that invested the most in the commodification of gay culture. In this respect, gay villages represent cosmopolitan environments which aim to attract tourists and affluent residents.

As commodification in the late post-industrial phases of advanced capitalism can incorporate some previously marginalised groups, herein gays, for example through processes of commercial and residential gentrification and media marketization, some gays and lesbians remain excluded from such villages, frequently on the basis of their seemingly extravagant gender identity, race, age or class (Binnie, Skeggs, 2004; Rushbrook, 2002). Therefore it is not a surprise that recent commodification also appears to cause segmentations among gay people, while constructing what should be the 'good gayness' and diffusing this model of the 'global gay' in the 'West' and worldwide (Altman, 2001). According to Bell and Binnie (2004), this is connected to the process of 'homonormativity' and to a broader agenda of assimilationist sexual citizenship, producing a global repertoire of themed gay villages.

Consequently, all the labels proposed as an interpretation of the first component (comfort, tourism, urban hierarchy) may be explained by the commodification of gayness, because comfort is linked to tourism which may be a result of the urban marketing of gay villages. As it is, these gay villages

have developed in major 'Western' cities as homosexuals and gays moved to large cities to escape historical sexual and social constraints of traditional life (Aldrich, 2004), and in some instances (North American cities) to gain political power by residential concentrations in former 'gay ghettos' (Castells, 1983). Since the 1980s, several cities in North America, Western Europe or Australia, became the receiving zones of global capitalist flows (Lauria and Knopp, 1985), due to local governments' actions for urban marketing and the gentrification of gay neighbourhoods. In addition, a commercial gentrification is at stake in several cities worldwide, including events that cater for gay people such as 'gay and lesbian prides' or 'gay games' (Wait, 2006). These developments could be understood as a consequence of the rise of the 'recreational city paradigm' in some parts of the world. Beyond this theoretical background, a map (Fig. 3) of the CP1 scores may be useful for a better understanding of the PCA results. Several cities have positive scores on CP1, meaning that gay facilities like hotels or restaurants are over-represented in such cities. We have distinguished four main categories here.

The first group of cities comprises the national capitals such as Berlin, Paris, London, Madrid, Amsterdam, Brussels, Prague, Rome, Budapest, Copenhagen, Tallinn and Ljubljana. The second group comprises other cities with a minimum population of 500,000 inhabitants (Hamburg, Cologne, Nuremberg, Zurich, Milan, Barcelona, Antwerp, Edinburgh, Florence, Venice and Nice). The third group includes other prominent cities (Montpellier, Bologna10, Frankfurt, Pisa, Vigo, Dresden, Bruges and Charleroi), and the fourth group represents the seaside resorts (Las Palmas de Gran Canaria, Blackpool, Sitges, Mykonos, Ibiza or Viareggio). All of these four groups of cities obtained significantly positive scores on CP1.

In contrast to these positively represented groups, another group of national capital cities is slightly negative, so they are under-equipped (Dublin, Lisbon, Oslo, Stockholm, Helsinki, Athens and Vilnius) or heavily under-equipped (Luxemburg, Riga, Bratislava, Bucharest and Sofia) with respect to this gay commodification index. In this latter grouping, two subgroups can be discriminated. First, cities that are located in rather isolated but rich regions with long-term effects of commodification (Dublin, Oslo and Helsinki), and secondly cities which are less affluent and/or isolated when it comes to the timing of effects of commodification and discourses of capitalism coming from the West (Athens, Vilnius, Riga, Bratislava, Bucharest and Sofia). Luxemburg remains complicated, with a role played both by the small population size of the city and the presence of other large gay cities nearby.

Evaluating the first component scores map confirms, by and large, our interpretation of the first axis as a dimension of commodification and subsequent tourism activity. High and low scores on the first component, however, are surely not only influenced by the impacts of capital, but also by a wider array of cultural effects such as religious and traditional values (Stulhofer, Rimac, 2009), or the transitions towards more post-materialist cultural values (Inglehart 2006) - such as liberal values, civil rights and democratization. Indeed, the wide array of research conducted during the past two decades suggests that this cultural background plays an

9 Data source: City tourism statistics compiled by European Cities Tourism (2011). Data for cities such as Paris, London, Amsterdam or Vienna were not available.

10 Bologna is often portrayed as city with the highest number of gays in Italy. See for example p.17 of the following report: http:// www.creativeclass.com/rfcgdb/articles/Italy%20in%20the%20Creative%20Age.pdf

Fig. 3: CP1 scores for selected cities in Europe

Sources: Spartacus Gay Guide (2007); made with Philcarto: http://philcarto.free.fr

important role in the development of gay places. We will elaborate these factors in more detail while uncovering how they are related to the second component.

5. Locating places of gay sociability and problematizing spatial gay (in)visibility

This section discusses and interprets the second component (CP2) on the basis of the component loadings (i.e. the correlations between variables and the component: see Fig. 2 above), as it would be difficult to interpret the second component's map prior to analysing these loadings. In such an attempt, the highest correlations will be given priority.

Gay bars have the highest positive correlations (+ 0.63) on CP2 and thus constitute the best variable for interpreting this component. On the opposite side, it is 'gay/tourist info' (places where gay information may be found: - 0.58), gay blue movies (gay porn cinemas: - 0.5) and health groups (- 0.45), which have the highest negative correlations. A comparative reading of Fig. 4a (map of CP2 scores) and of Fig. 4b (map of bars) makes their relationship clear from their common distribution patterns.

Gay bars are portrayed by several authors as sociability places11 for gays (see Marianne Blidon for France, 2007), or as Salinas (2007) puts it for Spain, spaces for their social relationships. Gert Hekma (2006) underlines the bars' exclusivity to the gay scene of the 1960s and the 1970s: that

is to say, a gay scene without gay bars was then unthinkable. Bars also played a crucial role in the rise of the gay movement in the USA (cf. the Stonewall Inn). Opening a gay bar was seen as both a militant and an entrepreneurial act at the same time by the owners of gay bars, as several authors have pointed out for Paris's gay village "le Marais" (Sibalis, 2004; Giraud, 2009). These gay businesses spearheaded commercial gentrification in many major cities in the "West" (Western Europe and North America mainly) and contributed to subsequent gay visibility.

Nowadays, it is very common in several West European cities to see gay activist activities linked to a particular bar (associative bars). Furthermore, gay bars serve as places of social (re)production of gay sub-cultural values of masculinity, as they differentiate on the basis of their performed gay identities, especially in major cities (Bell, Valentine, 1995). In this sense, bars contribute to a long-term construction of gay identities by giving room to its expression, performativity and reproduction for some gay people (De Busscher, 2000). Gay bars moreover cater to other people including bisexuals, or 'interested straights', in contrast to venues presenting movies with explicit sexual content (blue movies12). These cinemas cater exclusively for men. Showing 'blue movies' puts these cinemas into the category of 'places of sexual encounters', that Salinas (2007) described in the Spanish case to be places of furtive (sexual) encounters that are characteristic for places of reduced

11 Sociability places should be understood as places where sociability occurs, whereas social relationships spaces are those where social relationships can be built. We would like to point out here that gay bars may play both roles for gays and other sexual minorities.

12 These « Blue movies » also carry a 'VS' code for 'video shows' in the Spartacus Gay Guide.

homosexual visibility. Thus, we can conclude at this stage that the second axis contrasts closeted (hidden and more secretive) gay subculture with a more open and visible mainstream gay subculture.

The two categories 'gay/tourist info.' and 'health group' were also among the highest negatively correlated variables with respect to CP2. We will discuss them further in order to test the hypothesis about this axis (CP2).

The 'gay/tourist info.' variable is an aggregated category (Fig. 2, short label 9) and includes local gay groups and inquiry points for tourists. Fig. 5b shows that this category of gay facilities is present in every major Italian city in the sample, except for Viareggio and Venice. Gay groups seem to have a wide diffusion in Italian cities. Such an observation leads to several questions. Does this point to a real difference in Italian gay groups' capabilities (as opposed to the other

Fig. 4: a) City scores on the second component (PC2). Source: authors' calculations; b) Percentage of gay bars in the total number of 'gay places' per urban area in the EU, Norway and Switzerland in 2007. Sources: Spartacus Gay Guide (2007); made with Philcarto: http:Hphilcarto.free.fr

Fig. 5: a) Percentage of health groups in the total number of 'gay places' per urban area in the EU, Norway and Switzerland in 2007; b) Percentage of gay info. centres in the total number of 'gay places' per urban area in the EU, Norway and Switzerland in 2007. Sources: Spartacus Gay Guide (2007); made with Philcarto: http:Hphilcarto.free.fr

countries in the European Union), or is this evidence of an increased willingness to be more visible to their target audience? Could their higher number be a sign of a greater need for more in Italy? It should also be

considered that in Italy, gay groups are playing an equivalent role that associative gay bars did in other Western countries in the 1970s and 1980s, perhaps as a tool for the construction of gay visibility and acceptance in the wider society.

The Italian results are significantly different enough that we may interpret them by a larger set of arguments which applies also to discussion of CP1 results. We need to consider a broader field of the economic, cultural and societal factors that are involved here (Fig. 6). These factors should be discussed as background to our data (CP1 and CP2), as they reflect and impact the development of gay places.

Every nation, every region and every society has more or less different historical experience, some countries developed democracy later some sooner, some regions were rich in natural resources, etc. In short, some societies are more liberal about LGBT issues than others. Inglehart (2006) has connected this to the wider trend of transition from material (conservative, traditional and insecure) to post-material (more secure and liberal) values in societies. Interestingly, many LGBT sociological and activist reports agree (COE, 2011; FRA, 2008, 2009, 2010; EC, 2009; ILGA, 2013), and use societal values for explaining different attitudes14 in different regions. Without straying into deeper discussion of these issues, we should be aware of certain historical pre-conditions (history of law and medicine) connected with these attitudes towards non-heterosexuals: for example, how long homosexuality has been decriminalized in the country (Waaldijk, 2000), or when it stopped being considered as a disease by practitioners. The 'societal image of homosexuality' is impacted, hence it affects the 'strictness of heteronormativity'.

Other conditions, such as the prevalence of 'traditional religious, national and familial values' are also important.

The presence of strong orthodox religious discourse in a society may prefer certain heteronormative or discriminatory interpretations of sacred texts (COE, 2011). Similarly, some national discourses connected to 'ways of living your life as a proper citizen' may be connected to norms in sexuality, but also to 'accepted ways of family building', with an inclination to policing 'stereotypical' feminine and masculine gender roles in women and men. Moreover, if there is no anti-discriminatory legislation addressing sexuality and gender discrimination in the region, then even media, politicians or other opinion makers may (re)produce discourses which reinforce regional heteronormativity. This therefore impacts the security and comfort of non-heterosexuals. The structure of gay places may then be seen as a mirror to these conditions, combined with other wider regional economic and political conditions.

In the case of Italy, then, it is plausible to interpret the differences which are clearly in contrast to the other European countries, using some of these factors. Firstly, Italy has, to date, enforced only very limited anti-discriminatory jurisdiction (COE, 2011; ILGA, 2013). Secondly, the Vatican has a strong geographical position in the south of the country and in Rome, but its influence extends over local media as well. Thus far, neither the Catholic Church nor the Pope has showed any liberal progression in attitudes towards homosexuals. Moreover, societal reports (COE, 2011) rank Italy among the most traditional (in the foregoing sense) of European societies (together with Albania, Greece, Montenegro, Russia and Ukraine). On the other hand, as quite visible from the discussion of CP1, Rome is an important entrepreneurial capital (together with Milan, Florence, Venice and Bologna) and shows up as being more liberal. In general, we believe that it is therefore much less comfortable to be visibly 'gay' in an Italian city than in similar cities in Germany, UK or Spain, even if less comfort in the visibility of gayness does not necessarily prevent a person from finding other ways of performing one's gayness.

Fig. 6: The wider field of economic, cultural and societal factors behind the foundations of heteronormativity Source: authors'

13 LGBT is an acronym used to refer to lesbians, gays, bisexuals and transgender people.

14 For example, attitudes to discrimination of sexual minorities or towards LGBT people in general.

A recent study by Nadine Cattan and Alberto Vanolo (2011) is useful in addressing some of these issues. The authors evaluate lesbian night-time events in Paris and 'GLBTQ nights' in Turin (Italy), and explain part of the invisibility in the case of Italy as one of "ephemeral and temporal spaces". Such events "are performed in temporary places that are in clubs that, for the rest of the week, are intended for other customers and for different sexual identities [heterosexual]" (Cattan, Vanolo, 2011).

Furthermore, these authors described the Turin "GLBTQ scene" as follows: "...In Turin there are basically not 'fixed' GLBTQ discos and clubs, but just particular thematic nights (apart from three gay-friendly bars). Currently, mainstream GLBTQ events take place every Friday, Saturday and Sunday nights, organized by different associations, while minor events take form in residual and 'alternative' spaces. The spatial presence of the gay commercial spaces is therefore limited to certain times, and dispersed in the city space, so there is nothing like a 'gay area'."

Even though these gatherings are temporary, they act like spaces of empowerment, enabling LGBT people to dream and live reassuring emotions in the space and time period between the parties. This leads to the assumption that local Italian gays' sociability occurs in different ways, compared to other European cities, where gay districts (or gay villages) are often located in town centres and are therefore visible. These observations echo those of Luigi Mosca's15 about southern Italy - that local associative initiatives are more successful in touristic coastal areas when such initiatives can enjoy more visibility than in core areas. Therefore, in Italy, gay sociability occurs more in touristic areas like beaches, or in gay-friendly spaces like discos and clubs, characterised by greater degree of fluidity when it comes to the definitions of sexual identity.

Following the Spartacus Gay Guide, health groups give information in particular about AIDS. There are few European cities where such groups are mentioned. Interestingly, Italy has the highest number of cities that have such health information facilities. We are not able to explain this overrepresentation of health groups in Italy, however, apart from the previous questions raised about gay groups in Italy in general, such as the willingness of Italian gay groups for greater visibility.

The analysis of mapping the scores of the second component using the observed distribution patterns across the EU, Norway and Switzerland may now be discussed. This map (Fig. 4) generally sets German cities - regardless of their hierarchical position and where bars are overrepresented in the 'gay places' structure - in opposition to Italian cities, where gays bars are underrepresented. Most of the other seaside resorts also have negative scores on CP2, except for Las Palmas (Canary Islands, Spain). These values can be explained by an overrepresentation of cruising places (beaches) in these locations or by an underrepresentation of bars. Geographers, for example Emmanuel Jaurand and Stéphane Leroy (2008), have pointed to the role that Mediterranean beaches have played in homoeroticism and gay sociability, particularly as the cultural inheritance of Ancient Greece. This point strengthens an interpretation of the CP2 as a gay sociability component.

Some well-equipped capitals such as Prague, Budapest, Tallinn are negatively correlated with CP2, and this signifies that their commercial structure is less dominated by bars. Similar results with negative CP2 numbers can be found for Dublin and the other Western European cities, Scandinavian capitals, and for Athens, Bucharest, Vilnius and Warsaw. Finally, Bratislava, Sofia, Riga and Luxemburg, however, show an opposite situation, with slightly positive scores at CP2.

This can be interpreted as the limited visibility of homosexuality in these countries - either due to a greater acceptance of homosexuality in the wider society as in Scandinavia, UK or Benelux, or in contrast due to 'stricter heteronormativity' and the more conservative orientation of other countries (Slovakia, Bulgaria, or Latvia). Even in less accepting regions, however, where gay culture remains mostly invisible, non-heterosexual people are present and develop strategies for living out their sexuality socially. Anthropologist Liselotte van Velzen (2004), in her study on Belgrade, pointed out that such strategies include passing16 as straight or migrating to cyberspace17 to preserve anonymity. These strategies, however, maintain these homophobic economic and social environments largely uncontested.

In such situations, the Internet plays an important role as it enables communication between and within sexual minorities - and subsequently allows 'empowerment', since every otherwise spatially isolated individual can experience a (virtual) gay community and feel strengthened in their gay identity construction. The Internet can be viewed as well as an important factor influencing the very existence of gay places, such as venues, by reducing their number or making them invisible in the urban space. This hypothesis is corroborated by the findings of Brad Ruting (2008), who studied the economic transformation of gay spaces in Sydney, finding that the Internet is an important factor. This hypothesis may also partially explain why gay bars do not dominate cities and activist networks in Bulgaria or Romania, where local societies are rather 'traditional' - even though they are networked.

Generally speaking, the CP2 dimension may be interpreted as showing the local conditions of gay sociability (i.e. semi-public places where many forms of gay sociability may occur regardless of visibility). This finding calls for a discussion of gay (in)visibilities as they may also be proposed as an interpretation for the CP1 dimension, since it opposes cruising places (invisible) which have the highest correlation on it. Nonetheless, one should keep in mind that bars, as the penultimate indicator of gay visibility, have their best correlation on CP2, and not on the primary CP1 dimension.

6. Discussion: Challenging gay (in)visibilities in Europe

An important basis for discussion of the socio-spatial organization of gay lives is the European survey on discrimination in the EU, conducted in 2009 (European Commission, from now on EC 2009). Some results of this survey deserve a closer look with regard to the '(in) visibility' axis of 'gayness' in Europe, and thus to the interpretation of CP2.

15 MOSCA Luigi, Ph.D. Anthropology, Universita Degli Studi and Université de Bruxelles, interviewed on 02/08/2012

16 Shifting sexual identities, i.e. performing gay identity only with gay-friendly friends, while passing as straight or behaving inconspicuously when in public spaces.

17 Internet chat rooms, etc.

The 2009 survey underlines the relationships between gay invisibility and perceived discrimination towards LGBT people. It clearly shows that the lower the perceived discrimination in a country is, the lower is the percentage of people who have friends among homosexuals. This may be an illustration of a reaction where people who do not rate sexual discrimination as an important societal challenge, do not notice it and may even deny its existence as they do not have any LGBT friends or relatives. Consequently, this leads to lower perceived discrimination in ascending EU states (Bulgaria, Czech Republic (both at 22% of the surveyed citizens), Slovakia (27%) and Estonia (28%). Such results are surprisingly below the EU average (47%). Therefore, it can be suggested that the survey data are indicating the results of ignorance towards homosexuals rather than perceived discrimination.

The survey also describes a decrease in the perceived extent of discrimination throughout the EU (compared with a 2008 survey), but it is still seen as particularly widespread in many of the Mediterranean countries (Cyprus: 66%, Greece: 64%). Furthermore, Italy and France, both with 61%, show results far above the EU average of 47%. The report also underlines the situation in the Netherlands where discrimination on grounds of sexual orientation is thought to be growing. This development can be linked to an increased incidence of attacks against LGBT people there in recent years. At the same time, significantly fewer citizens in Italy and the UK now believe that 'discrimination on grounds of sexual orientation is widespread in their country' than what was recorded in 2008. If we examine the averages on the comfort scale from 1 to 10, the results of this survey reveal that people in Sweden (8.7), Denmark (8.4) and the Netherlands (8.2), are the most comfortable with having a LGBT leader in their country, while people in Bulgaria (3.2), Romania and Turkey (each 3.4) report feeling the least comfortable (EC, 2009).

Another study, conducted by Norwegian researchers (R0thing, Bang, 2010), explains how difficult it can be for LGBT youth to accept their homosexuality, even in a country as tolerant as Norway18. Disturbing linkages between homophobia (or heteronormativity) and suicidal behaviour in youngsters had been reported. The authors (ibid. 2010) explain this paradoxical situation as the result of the lack of "non-heterosexual future-scapes" offered in school and in society in general, creating young peoples' fear of being or becoming homosexual (homo-negativism). On the other hand, Norway has a marriage act that states the equality of homosexual and heterosexual relationships. The authors argue, however, that "many of those who support the law and the rights of lesbians and homosexuals do not necessarily find homosexuality as desirable or as worthy of being promoted as heterosexuality" (ibid. 2010: 160).

This 'homo-negativism', or maybe better 'strict heteronormativity', is occurring also in other European countries with egalitarian legislations like those in Scandinavian countries, and we believe that in part this may explain the remaining gay invisibility in some regions of these countries.

As outlined above, some wider field of economic, cultural and societal factors is involved, specifically that connected with 'traditionalism' and 'liberalism' in people's values, and we will extend these arguments a little further (see Fig. 6,

again). As in the case of Italy, the strong position of the Church seems to be important for the preservation of 'conservative and traditional' discourses. We are convinced from these data that a religion's position or opinion is of utmost importance with respect to gay visibility and the existence of gay spaces, in general, and not only across the EU.

Clearly, religion is most important here, but it would be incorrect to see it simplistically as an epitome of intolerance or even hate: religions as ideologies cannot be generally dismissed as intolerant or rejectionist. Clear differences may be thus identified in European countries, where the major religion is Christianity. First, Scandinavian countries together with most of the western European countries, are predominantly Protestant. Protestantism has been shown to adopt the most liberal or progressive attitudes towards homosexuality (Stulhofer, Rimac, 2009). Perhaps also due to this factor, it is possible for homosexuals to get married in churches there (e.g. in Denmark or Sweden). A variety of southern and central European countries are mostly Roman Catholic (e.g. Poland, Slovakia, Hungary, Spain, and Italy). Here Italy is additionally influenced by the position of the Vatican State, as discussed before. Lastly, most eastern European countries are predominantly Orthodox: perhaps the best example is Russia, where the contemporary worsening situation of LGBT rights illustrates not only the conservative attitudes of the Orthodox Church, but also its conjunctions to the state. In spite of the fact that Russia is not a member state of the EU, it has a profound impact on political and ideological development in most of the Eastern European region. Stulhofer and Rimac (2009) give good examples of these religious and cultural influences in more detail. Even though more religious countries show lower scores for gay presence (lower numbers of gay bars: see Fig. 4b) and visibility, causality between religious structure and homosexuality is much more complex and deserves further investigation. Despite these facts we conclude that the impact of religions is somewhat traditional and plausibly reactionary towards gay (in)visibility.

7. Greater gay acceptance in wider society

In presenting the results of cities on the CP2 dimension, we distinguished two groups of cities connected with negative CP2 results and therefore to gay (in)visibility. The first group was connected to invisibility largely because of the previously-mentioned factors connected to general heteronormativity (Bratislava, Sofia, Riga, etc.). For the second group (Scandinavian capitals, some Western European cities, Prague or Budapest), we implied that promotion of liberal values in their societies might cause lesser need for exclusive socializing in 'gay bars', hence less demand for them. This is an interesting 'trend' to elaborate.

Our own observations in Prague, Budapest and Copenhagen indicate that the development of gay visibilities does not necessarily show in the form of signposting gay symbolism, such as rainbow flags or stickers in front of businesses. In Prague, gay people are becoming more and more visible in spaces which are not labelled by 'gay exclusivity', and this may in turn lead to reducing the need to open a gay bar. As a result of this, the niche market which capitalizes LGBT clientele might also be weakened as a result of choice instead of necessity.

18 According to the ILGA-Europe Rainbow Europe Index 2013 report, Norway is second to the UK as the most sexually equal country in Europe.

We find it useful to juxtapose the progressive commodification strategies with slow but growing acceptance of LGBT people. The phenomenon of an increasing acceptance of a diversifying society, where the previous state was dominated by gay exclusivity, can be named 'de-gaying', and this holds especially true for Copenhagen. Danish people are one of the most tolerant in the EU (EC, 2009; EU LGBT survey, 2013). This country was, for example, the first to introduce registered partnerships in 1986, and had founded an LGBT organization originally in 1948. This process of 'gayness' integration in wider society now makes 'exclusive' (i.e. visible or marketed) gay clubs economically and socially obsolete. Despite the positive reading of this, the term 'de-gaying' is used by many authors in a rather negative way (Whittle, 1994; Collins, 2004), when they are addressing a wider assimilation of 'gayness' into a commoditized cosmopolitan culture, when consumable 'appropriate gayness' has become an urban spectacle (Rushbrook, 2002). It would be, therefore, naive to read the recent changes in gay visibilities as a result of the sole growing acceptance of gays and other sexual minorities.

Many cities in Central Europe, including Prague, have communist legacies, which at least to some degree succeeded in lowering the social differentiation process, consequently with the drawback of making homosexuality a taboo. Thus, no 'explicit' gay clubs existed during the communist era. Although the contemporary Prague 'gay scene' is slightly over-equipped (CP1), its future existence may eventually become unimportant, even when the commodification of 'gayness' is in its early phase there. On the one hand, new types of gay businesses are appearing (gay travel agencies, etc.) and on the other, these market potentials may diminish with the rise of societal acceptance. Therefore, the existence of gay venues can be seen as complex, and the commodification processes favouring capitalisation or exploiting the unequal situation of homosexuals have to be further problematized, especially at the present time of a continuous liberalization of cultural values.

Concluding the issue of gay (in)visibilities, we would like to include a short anecdote that took place in Prague. It represents a good example of what 'gay (in)visibility' can be and how problematic it is for researchers to measure it. On Christmas Day 2012, a family (mixed-sex parents with two children) visiting Prague, entered a gay men bear19 bar for a moment, while not knowing exactly what kind of bar it was and then leaving the place after realising that it was a gay bear bar and not a place they intended to visit. This short experience of less than two minutes might be considered as a proof of gay (in)visibility in Prague, because this bar is easily accessible to everyone. At the same time, one may argue that bar was not 'visibly gay enough' in the public space, and that this is the reason why the family did not initially hesitate to enter. Furthermore, the tiny rainbow sticker on the front door was probably not seen or interpreted as an indication of an LGBT venue. So, if gay symbols are only recognizable by gay people and not by the wider population, this may to certain extent explain the somewhat still limited co-existence of heterosexual and non-heterosexual cultures. This experience may indicate that gay facilities' (in)visibility needs to be studied from several perspectives, including direct observations (e.g. crowded places, or not from outside, presence/absence of rainbow flags) and coming from gay

patrons' and non-patrons' perspectives and from straight people's perspective. This cross-perspective analysis would be helpful to describe gay (in) visibilities in Europe beyond their general contexts.

8. Conclusions

Quantitative methodology enables researchers to look beyond simple or multiple indicator data. We believe that we have shown that, even in a field largely dominated by qualitative methods, the geographies of sexualities can be studied by quantitative methods. Quantitative tools do not have to strive for generalizability or universal validity, but they can complement other methodologies in dialectical paths to understanding.

Principal Component Analysis (PCA) was used for an analysis of data from the Spartacus International Gay Guide of 2007, which made it possible to study gay facility structures and places in the largest cities of the European Union, Norway and Switzerland. We wanted to understand the nature of the underlying economic, cultural and social factors which affect the existing structure of various European 'gay places'. PCA has helped in reaching beyond the 'factual data', producing two new components (CP1 and CP2) with characteristic functions which allowed for a further investigation of this underlying structure. The first component (CP1) has been interpreted as a comfort, touristic or commodification axis, whereas the second component (CP2) shed some light on the complex conditions of gay sociability, and specifically on gays' cultural (in) visibility, taking the touristic or commodification dimension - as accounted for by the first component - into account.

This study offers some explanations for gay (in)visibilities in Europe, since it might be also a valuable interpretation for CP2, as gay sociability may occur visibly and invisibly. Plausible explanatory factors for CP1 and CP2 include historical experience and political background (LGBT legislation, such as (de)criminalization or protection against the economic situation, market economy, commodification, tourism, etc.), and other societal or cultural conditions (traditional vs. liberal values). These factors have to be understood in relation to other factors influencing broader concepts such as religion, nationalism, etc., or mirroring them through media, politics and opinion makers. It is then a whole 'web' of factors that affect levels of gay acceptance, and gay cultural development either in space-time or in the cyberspace.

We believe that the PCA methodology used in this research has shed some light on the distribution of gay spaces and places, but we have to be aware of some connected analytical limitations. The two components examined account for some of 57% of the variance in the original set of 27 variables (i.e., the information available from the data set), and therefore some important factors still remained hidden to us. In order to describe gay facilities' visibility in more detail, we suggest that conducting a multi-actor (residents, tourists, activists, city planners, municipalities, etc.) qualitative or mixed analysis should be designed. This quantitative study of 'gay places' in the largest European cities focused mainly on a national and macro-regional analytical perspective, which could be complemented in the future with smaller-scale surveys. Our main goal was to utilize a standardized database which allowed for a

19 Though a closed definition of « bears » does not exist, the term can be said generally to refer to gay or bisexual men with a good deal of body hair (Textor, 1999).

comparative viewpoint. Even though this study explicitly focusses only on gay males and venues catering to them and was not meant to engage in a refined critical discussion of queer theory, the authors believe that the results and issues from this discussion are of a character that also influence cultural attitudes towards homosexual women and the other non- heterosexual people in general.

Acknowledgement

This work was supported by research projects 'SVV No. 260082 and GA UK No. 1092214 „Queer geography: Commodification, segregation and globalization of sexual identities in cities?" at the Department of Social Geography and Regional Development, Charles University, Prague.

References:

ADLER, S., BRENNER, J. (1992): Gender and space: lesbians and gay men in the city. International Journal of Urban and Regional Research, No. 16, p. 24-34. ALDRICH R. (2004): Homosexuality and the city: an historical

overview, Urban Studies, Vol. 14, No. 9, p. 1719-1737. ALTMAN D. (1996): Rupture or continuity. The internationalization of gay identities. Social text, No. 48,

p. 77-94.

ALTMAN D. (2001): Global sex. Chicago and London: The University of Chicago press. 192 pp.

BASSI, C. (2006): Riding the dialectical waves of gay political economy: a story from Birmingham's commercial gay scene. Antipode: a radical journal of geography, Vol. 38, No. 2, p. 213-235. BELL, D., BINNIE, J. (2004): Authenticating Queer Space: Citizenship, Urbanism and Governance, Urban Studies, Vol. 41, No. 9, p. 1807-1820. BELL, D., VALENTINE, G. [eds.] (1995): Mapping desire: Geographies of sexualities. Psychology Press. London: Routledge, 384 pp. BINNIE, J., VALENTINE, G. (1999): Geographies of sexuality: a review of progress. Progress in Human Geography, Vol. 23, No. 2, p. 175-187. BINNIE, J., SKEGGS, B. (2004): Cosmopolitan knowledge and the production of sexualized space: Manchester's gay village. The Sociological Review, Vol. 52, No. 1, p. 39-61. BLIDON, M. (2007) : Ville et homosexualité, une relation à l'épreuve de la cartographie. In : Pumain, D., Mattei, M.-F. [coord.]: Données urbaines, t.5. Economica-Anthropos, Paris, p. 67-76. BONDI, L. (1998): Gender, class, and urban space: Public and private space in contemporary urban landscapes. Urban Geography, Vol. 19, No. 2, p. 160-185. BROWN, M., KNOPP L. (2002): Queer Cultural Geographies: We're Here! We're Queer! We're Over There, Too! In: Anderson, K., Domosh, M., Pile, S. [eds.]: The Handbook of Cultural Geography. London, Sage, p. 313-324. BROWN, M., KNOPP L. (2006): Places or polygons? Governmentality, scale and the Census in The Gay and Lesbian Atlas, Population, Space and Place, No. 12, p. 223-242.

BROWNE, K., LIM, J., BROWN, G. (2007): Geographies of Sexualities. Ashgate Publishing Limited, Burlington, 278 pp.

BROWNE, K. (2010): Queer quantification or queer(y) ing quantification, In: Browne, K., Nash, C. J., [eds.]: Queering Methods and Methodologies: Intersecting Queer Theory and Social Science Research. Ashgate, Farnham, p. 231-249.

BROWNE, K., NASH, C. J., HINES, S. (2010): Introduction: towards trans geographies. Gender, place and culture, Vol. 17, No. 5, p. 573-577.

BUTLER, J. (1993): Bodies that Matter: On the Discursive Limits of 'Sex'. New York, Routledge, 304 pp.

BUTLER, J. (1990): Gender Trouble. Feminism and the Subversion of Identity. New York, Routledge, 272 pp.

CASTELLS, M. (1983): The city and the grassroots, Berkeley, (CA). University of California Press, p. 138-170.

CATTAN, N., VANOLO, A. (2011): The struggle to belong. Dealing with diversity in 21st century urban settings, paper called "Homosexuality and the city: emotional geographies of clubbing in Paris and Turin" presented at the International RC21 conference, Amsterdam, July 2011, p. 1-22 [online] [cit. 07.02.2014]. Available at: http://www.rc21.org/conferences/amsterdam2011/ edocs2/Session%2011/11-1-Cattan.pdf

CATTAN, N., CLERVAL, A. (2011): Un droit à la ville. Réseaux virtuels et centralités éphémères des lesbiennes à Paris. Justice Spatiale, No. 3, p. 1-19.

COLLINS, A. (2004): Sexual Dissidence, Enterprise and Assimilation: Bedfellows in Urban Regeneration. Urban Studies, Vol. 41, No. 9: p. 1789-1806.

COOKE, T., RAPINO, M. (2007): The migration of partnered gays and lesbians betwen 1995 and 2000. Professional Geographer, No. 59, p. 285-297.

Council of Europe (2011): Discrimination on grounds of sexual orientation and gender identity in Europe. 2nd edition. Council of Europe Publishing, 132 pp. [online]. [cit. 07.02.2014]. Available at: http://www.coe.int/t/ Commissioner/Source/LGBT/LGBTStudy2011_en.pdf

DE BUSSCHER, P.-O. (2000): Le monde des bars gais parisiens : différenciation, socialisation et masculinité. Journal des anthropologues, No. 82-83 [online]. [cit. 02.08.2012]. Available at: http://jda.revues.org/3372

DOAN, IP L. (2007): Queers in the American city: Transgendered perceptions of urban space. Gender, Place and Culture, Vol. 14, No. 1, p. 57-74.

DUGGAN, L. (2003): The Twilight of Equality?: Neoliberalism, Cultural Politics, and the Attack on Democracy. Boston, Beacon Press, 136 pp.

DUNCAN, N. (1996): BodySpace: Destabilizing geographies of gender and sexuality. Routledge, London, 278 pp.

European Spatial Planning Observation Network, Study on Urban Functions (Project 1.4.3), Final Report, Chapter 3, ESPON, 2007 [online]. [cit.06.02.2014] Available at: http://www.espon.eu/export/sites/ default/Documents/Projects/ESPON2006Projects/ StudiesScientificSupportProjects/UrbanFunctions/fr-1.4.3_April2007-final.pdf

European Union Agency for Fundamental Rights (2008): Homophobia and Discrimination on the grounds of Sexual Orientation and Gender Identity in the European Union Member States: Part I - Legal Analysis, 2008 [online]. [cit.07.02.2014] Available at:

http://fra.europa.eu/en/country-report/2012/country-reports-homophobia-and-discrimination-grounds-sexual-orientation-part-1

European Union Agency for Fundamental Rights (2009): Homophobia and Discrimination on Grounds of Sexual Orientation and Gender Identity in the EU Member States: Part II - The Social Situation, 2009, 157 p. [online]. [cit.14.10.2013] Available at: http://fra.europa. eu/sites/default/files/fra_uploads/397-FRA_hdgso_ report_part2_en.pdf

European Union Agency for Fundamental Rights (2011): Homophobia, Transphobia and Discrimination on Grounds of Sexual Orientation and Gender Identity: 2010 Update - Comparative Legal Analysis" [online]. [cit.07.02.2014] Available at: http://fra.europa. eu/sites/default/files/fra_uploads/1286-FRA-LGBT-report-update2010.pdf

European Union Agency for Fundamental Rights (2013): EU LGBT survey- European Union lesbian, gay, bisexual and transgender survey, results at a glance, 31 pp. [online]. [cit.06.09.2013] Available at: http://fra.europa.eu/sites/ default/files/eu-lgbt-survey-results-at-a-glance_en.pdf

FERNANDEZ SALINAS, V (2007): Visibilidad y Escena Gay Masculina en la Ciudad Española. Documents D'analisi Geografica, No. 49, p. 139-160.

FOUCAULT, M. (1980): Power/Knowledge. New York, Pantheon Books, 288 pp.

FOUCAULT, M. (1995): Discipline and Punish: The Birth of the Prison. New York, Vintage Books, 333 pp.

GIRAUD, C. (2009): Les commerces gays et le processus de gentrification: l'exemple du quartier du Marais à Paris depuis le début des années 1980. Métropoles, No. 5, p. 75-115.

GORMAN-MURRAY, A.; BRENNAN-HORLEY, C., MC LEAN, K., WAITT, G., GIBSON, C. (2009): Mapping same-sex couple family households in Australia. Journal of Maps, Vol. 6, No. 1, p. 382-392.

GORMAN-MURRAY, A., WAITT, G., JOHNSTON, L. (2008): Guest Editorial - Geographies of Sexuality and Gender 'Down Under'. Australian Geographer, Vol. 39, No. 9, p. 235-246.

GRUSZCZYNSKA, A. (2007): Living 'la vida' Internet: Some Notes on the Cyberization of Polish LGBT Community [online] [cit. 11.09.2012]. Available at: http://www.policy.hu/takacs/books/isbn9616455459/pdf/ peacemirror/07Gruszczynska.pdf

HEKMA, G. (2006): Le monde gay, de 1980 à nos jours. In: Aldrich, R. [dir.]: Une histoire de l'homosexualité. Seuil, Paris, p. 333-363.

HEMMINGS, C. (2002): Bisexual spaces: A geography of sexuality and gender. Psychology Press, Routledge.

HOWELL, IP (2009): "Sexuality". In: Kitchin, R., Thrift, N. [eds.], International Encyclopeadia of Human Geography. Oxford, Elsevier, pp. 119-124.

HUBBARD, P (2001): Sex zones: intimacy, citizenship and public space. Sexualities, Vol. 4, No. 1, p. 51-71.

HUGHES, H. L. (2006): Pink Tourism: Holidays of Lesbians and Gay Men, CABI, 256 pp.

HUGHES, H. (1997): Holidays and homosexual identity. Tourism Management, Vol. 18, No. 1, p. 3-7.

ILGA-Europe (2013): Annual Review of the Human Rights Situation of Lesbian, Gay, Bisexual, Trans and Intersex People in Europe [online]. [cit.07.2.2014] Available at: http://www.ecoi.net/file_upload/90_1369137411_ilga-europe-annual-review-2013.pdf

INGLEHART, R. (2006): Mapping global values. Comparative Sociology, Vol. 5, No. 2-3, p. 115-136.

JAGOSE, A. (1996): Queer Theory, An Introduction. New York University Press, New York, 153 pp.

JAURAND, E., LEROY, S. (2008): Le littoral, un paradis gay? Proceedings of the international pluridisciplinary conference "The littoral: challenge, dialogue, action", Lille, France, 16-18 January 2008, p. 1-13. [online] [cit. 07.02.2014]. Available at: http://www.meshs.fr/ documents/pdf/publications/actes/colloque_littoral/ Jaurand-Leroy.pdf

JOHNSTON, L., VALENTINE, G. (1995): Wherever I lay my girlfriend, that's my home: the performance and surveillance of lesbian identities in domestic environments. In: Bell, D., Valentine, G. [eds.]: Mapping Desire geographies of sexualities. Routledge, London, p. 99-113.

JOLLIFFE, I. T. (2002): Principal Component Analysis, Second Edition, Springer-Verlag, New York, 518 pp.

KIRBY, S., HAY, I. (1997): (Hetero)sexing space: gay men and "straight" space in Adelaide, South Australia. Professional Geographer, Vol. 49, No. 3, p. 295-305.

KNOPP L. (1987): Social theory, social movements and public policy: recent accomplishments of the gay and lesbian movements in Minneapolis, Minnesota. International Journal of Urban and Regional Research, Vol. 11, No. 2, p. 243-261.

LAURIA, M., KNOPP L. (1985): Toward an analysis of the role of gay communities in the Urban Renaissance. Urban Geography, Vol. 6, No. 2: p. 152-169.

LEROY, S. (2005): Le Paris gay. Eléments pour une géographie de l'homosexualité. Annales de Géographie, No. 646, p. 579-601.

LEVINE, M. (1979): Gay Ghetto. In: Levine, M.: Gay Men: The Sociology of Male Homosexuality. Harper & Row, New York, p. 182-204.

MILLER, V. (2009): Gay Geographies. In: Kitchin, R., Thrift, N. [eds.]: International Encyclopeadia of Human Geography. Elsevier, Oxford, Vol. 4, p. 302-308.

MORAN, L., SKEGGS, B., TYRER, P, CORTEEN, K. (2001): Property, boundary, exclusion: making sense of hetero-violence in safer spaces. Social & Cultural Geography, Vol. 2, No. 4, p. 407-420.

NASH, C. J. (2010): Trans geographies, embodiment and experience. Gender. Place and Culture, Vol. 17, No. 5, p. 579-595.

OSWIN, N. (2008): Critical geographies and the uses of sexuality: deconstructing queer space. Progress in Human Geography, Vol. 32, No. 1, p. 89-103.

Philcarto programme [online]. [cit. 27.05.2013]. Available at: http://philcarto.free.fr

PRATT, G. (2009): Heteronormativity. In: Gregory, D. et al. [eds.]: The Dictionary of Human Geography, 5th Edition. Wiley-Blackwell, Oxford, p. 329-330.

ROSE, G. (1995): Geography and gender, cartographies and corporealities. Progress in Human Geography, Vol. 19, No. 4, p. 544-548.

ROTHENBERG, T. (1995): 'And She Told Two Friends': Lesbians Creating Urban Social Space. In: Bell, D., Valentine, G. [eds.]: Mapping Desire Geographies of Sexualities. Routledge, London and New York, p. 150-165.

R0THING, Ä., SVENDSEN, S. H. B. (2010): Homotolerance and Heterosexuality as Norwegian Values, Journal of LGBT Youth, Vol. 7, No. 2, p. 147-166.

RUSHBROOK, D. (2002): Cities, queer space, and the cosmopolitan tourist. GLQ, Vol. 8, No. 1-2, p. 183-206.

RUTING, B. (2008): Economic Transformations of Gay Urban Spaces: revisiting Collins' evolutionary gay district model. Australian Geographer, Vol. 39, No. 3, p. 259-269.

SIBALIS, M. (2004): Urban Space and Homosexuality: The Example of the Marais, Paris' 'Gay Ghetto'. Urban Studies, Vol. 41, No. 9, p. 1739-1758.

Special EUROBAROMETER 317, Discrimination in the EU in 2009, Report: European Commission, November 2009, p. 85-91.

STULHOFER, A., RIMAC, I. (2009): Determinants of Homonegativity in Europe. Journal of Sex Research, Vol. 46, No. 1, p. 24-32.

TEXTOR, R. A. (1999): Organization, Specialization, and Desires in the Big Men's Movement: Preliminary Research in the Study of Subculture-Formation. Journal of Gay, Lesbian, and Bisexual Identity, Vol. 4, No. 3, p. 217-239.

VALENTINE, G. (1993a): Negotiating and managing multiple sexual identities: lesbian time-space strategies. Transactions of the Institute of British Geographers, New Series, Vol. 18, No. 2, p. 237-248.

VALENTINE, G. (1993b): (Hetero)sexing space: lesbian perceptions and experiences of everyday spaces. Environment and Planning D: Society and Space, Vol. 11, No. 4: p. 395-413.

VALENTINE, G. (2007): Theorizing and researching intersectionality: a challenge for feminist geography. The Professional Geographer, Vol. 59, No. 1, p. 10-21.

VAN VELZEN, L. (2004): Down and Out in Belgrade: An Ethnographic Account on the Everyday Life Experiences of Serbian Gays and Lesbians [online]. [cit. 11.09.2012]. Available at: http://www.policy.hu/takacs/books/ isbn9616455459/pdf/peace-mirror/02_van_Velzen.pdf

WAALDIJK, K. (2000): Civil Developments: Patterns of Reform in the Legal Position of Same-Sex Partners in Europe. Canadian Journal of Family Law, p. 62-88.

WAITT, G. R. (2006): Boundaries of desire: becoming sexual through the spaces of Sydney's 2002 gay games. Annals of the Association of American Geographers, Vol. 96, No. 4, p. 773-787.

WEIGHTMAN, B. (1981): Commentary: towards a geography of the gay community. Journal of Cultural Geography, Vol. 2, No. 1, p. 106-112.

WHITTLE, S. (1994): Consuming differences: the collaboration of the gay body with the cultural state. The Margins of the City: Gay Men's Urban Lives, p. 27-41.

Authors' addresses:

Koessan GABIAM, M.Sc.

Laboratoire de géographie humaine, Université Libre de Bruxelles (Belgium) Bâtiment D, 6ème niveau, campus du Solbosch CP130/03, avenue F.D. Roosevelt 50, 1050 Bruxelles, Belgium e-mail: koessan.gabiam@gmail.com

RNDr. Michal PITONAK

Department of Social Geography and Regional Development Faculty of Science, Charles University in Prague Albertov 6, 128 43 Praha 2, Czech Republic e-mail: pitonak@natur.cuni.cz

Initial submission 3 October 2013, final acceptance 25 March 2014 Please cite this article as:

GABIAM, K., PITONAK, M. (2014): Economic, cultural and social factors influencing the development of gay businesses and places: Evidence from the European Union. Moravian Geographical Reports, Vol. 22, No. 3, p. 2-17. DOI: 10.2478/mgr-2014-0014.