Determinants of the spatial diffusion of Bollywood cinema

Bollywood is one of many cinematographic centres in India. It has its seat in Mumbai, where most of its movies are shot. The centre produces approximately 200 movies per year, with nearly 2.5 million people working on them, which makes Bollywood the biggest movie centre in the world (Stachówna & Piekarski 2009). In this publication the notion of “Bollywood cinema” is taken to mean all the movies produced at least in part in Mumbai, and with Hindi as the prevailing language. At the turn of the 21^st century, there were many reports regarding Bollywood cinema and its diffusion to selected countries, whose authors were: A. G. Roy (2012), A. G. Roy and C. B. Huat (2012), R. Kaur and A. J. Sinha (2005) and R. B. Mehta and R. Pandharipande (2010). These works provide detailed analyses of selected productions and the history of Bollywood. However, no attempts have been made to determine the conditions in which Bollywood cinema was popularised all around the world. This paper is a summary of the key determinants that contributed to the diffusion of Bollywood cinema in all the countries in which these movies were distributed. In order to achieve the goals of the paper, an analysis was made of the distribution of all Bollywood films within a specific time frame – from 1970 to 2010. The analysis was carried out in an intervallic manner, so four full decades were used. The year 1970 was chosen as the beginning, because before that date the cinema distribution of Bollywood films in the world was sporadic and limited to only a few titles. It is worth specifying that the subject of the work is only Bollywood films with official cinematic distribution. Other media (DVD, VOD, YouTube and others) that are also popular, but equally difficult to measure and subject to reliable analysis, have not been included. For similar reasons, the focus was only on films produced in Mumbai and in Hindi or with Hindi as the dominant language. They constituted a large group of over 5,800 films.

As an element of Hägerstrand’s theory, the “neighbourhood effect” could play a key role in the diffusion of popular Hindi cinema, so the assumptions of his concept were used in the work.

The commonly accepted definition by T. Hägerstrand (1967) says that “[t]he diffusion of innovations – the origin and dissemination of cultural novelties – is an area of study which concerns all sciences dealing with human activity, including, not least of all, cultural and economic geography,” (1967 p. 1). Thus, diffusion of innovation theory tries to explain why and how this process of disseminating a new technological idea takes place. T. Hägerstrand (1967) isolated a neighbourhood effect and a wave model of spatial diffusion, among others. The first concerns the high probability of a community accepting an innovation when a neighbouring community has already accepted it. The wave theory compares the diffusion of innovation to circular ripples on water that are formed when a stone is thrown into it. They spread concentrically, starting with the source of information until the moment they reach the outer radius. For the innovation to diffuse according to the model it is necessary that it proceed in a perfectly homogeneous environment and not be disturbed.

The aim of information regarding a diverse group of countries (Pakistan, Nepal, Bangladesh, South Africa, Australia, Soviet Union [Russia since 1991], and Senegal) is to highlight the importance of other circumstances in this process that contributed to changing the course of the spatial diffusion. This research has also been prompted by the spectacular phenomenon of Bollywood cinema in countries distant both spatially and culturally from India that was observed in the 1^st decade of the 21^st century.

According to the directory of movies made in Bollywood in the last hundred years of this cinematographic centre’s operation, 5,832 movies were made there within the years 1970–2010. All of them have been analysed here.

The concept of spatial diffusion, as well as the nature of an analysed innovation, allowed the following hypotheses to be proposed:

To achieve the research objectives the following test procedure was carried out:

In geographical research of innovation diffusion, emphasis should be put on the function of the distance of the subjects spreading the innovation. Thus, the key dependent was the distance from Mumbai to the largest urban centre of the countries in which at least one premiere of a Bollywood movie took place. The classification criterion was an equal number of countries in each class – all the countries were divided into six groups, four of which included 13 countries, and two included 12 countries, and these were listed in order of increasing distance (Table. 2).

In the 1970s Bollywood movies were shown in the cinemas of 10 countries outside India. One belonged to class I, situated the nearest to the source of the innovation. From 1970 to 1979 showings were extended to three countries of class III, one of class IV, two of class V, and three of class VI. In the next decade there were 15 new countries, two of which were of class I, three – II, one – III, three – IV, five – V, and one of class VI. In the 1990s the number of innovation supporters rose by 11 countries: one of class II, two – III, three – IV, two – V, and three of class VI. The decade from 2000 to 2010 was characterised by a rise in the production and export of Bollywood movies, and these appeared for the first time in 10 countries from class I, 9 – II, 7 – III, 6 – IV, 4 – V, and 6 of class VI.

The advancement of Bollywood cinema’s spatial diffusion was measured by the number of movies distributed in the study period (Table. 1). For better clarity, the number of distributed movies throughout the whole period was averaged for each class (Table. 3). In the study period, in 13 countries of class I there were 20 movies distributed on average (max. 192 in Kuwait). In the countries of class II the average number of movies was 7 (max. 31 in Singapore). In the next group the average number of movies was 4 (max. 11 in Poland). In class IV the average was 6 and the leading country was Germany (24). Class V (average 53) includes Great Britain, where the biggest number of Bollywood movies of all the countries was distributed (439). In the countries of class VI the average number of movies shown was 25 (max. USA – 186).

Based on the collected information the conclusion can be drawn that the physical distance between Mumbai and the reception countries did not have an impact on either the timing of the premiere of Bollywood movies in these countries, or on the number of premieres. Thereby, in case of the spatial diffusion of Bollywood cinema, the assumptions of T. Hägerstrand’s theory regarding the significance of the relation that the course of the process has with the distance from source sites to reception sites were not confirmed. The aforementioned analysis allows us to confirm the first and second hypotheses proposed above. However, it is worth mentioning that for some groups of countries the neighbourhood effect can be observed. It is manifested by an interest in Bollywood productions in countries close to a country where they have already gained popularity. This applies primarily to European countries in the 1^st decade of the 21^st century, when the popularity of Bollywood productions spread across the borders of Great Britain to other countries, such as: Ireland, the Netherlands, France, Germany and Poland.

The popularity gained by Bollywood cinema at the turn of the 21^st century might have resulted from the Indian diaspora supporting their national cinema. This dependency is particularly evident from the mid-90s, when a tendency to screen the life of Indian emigrants emerged, accompanied by the image of Indian people living abroad becoming more positive. These steps contributed to the rise in audience numbers of spectators in cinemas among the Indian diaspora.

The countries where Bollywood movies were distributed were ranked by size of Indian diaspora. They were divided into groups of the same size (four classes of 13 countries and two of 12 countries; Table. 4).

In the period 1970–1979 Bollywood movies were distributed in 10 countries. In the next decade cinema repertoires were extended to include Bollywood productions in 13 countries, in 11 countries in the 1990s, and in 42 after 2000.

The advancement of the spatial diffusion in the countries ranked by size of Indian diaspora was shown using the average number of productions distributed in the countries of each class (Table 5). In class I countries, the average number of premiere screenings was 79 (439 in Great Britain, 192 in Kuwait, 186 in the United States). In class II countries, with a considerable size of Indian diaspora, the average was 12 (max. 42 in Australia). In class III, which has an average of 7, Germany stands out with 24 productions. In class IV the average number of movies screened in cinemas was 4 (the most being 13, in Japan). In the countries belonging to the two lowest classes there were few movies screened – in class V their average number was 10, due to the very large contribution of Ireland (99), and the considerable contribution of Poland (11). Among the countries from class VI, where the average was 2, the only countries that achieved better-than-average results were Turkey (6) and Peru (3).

The hypothesis regarding how the point in time that Bollywood movies were introduced to receptive countries’ cinemas relates to the size of their Indian diaspora was not confirmed. Bollywood cinema appeared in reception countries in different periods and it is hard to associate it with the size of diaspora. But such a dependency can be seen in how the number of Bollywood movies distributed in the reception countries related to the size of the Indian diaspora. The average number of distributed movies decreases in direct proportion to the number of people of Indian origin living in the groups of countries. These dependencies are not rectilinear – a distinct aberration can be observed in the group of countries that contains Ireland, where Bollywood cinema gained popularity despite there being only a small Indian diaspora. The reasons for this can be detected in the small distance from Great Britain, where the cinema was popularised due to its strong Indian diaspora. It is possible that this was due to the neighbourhood effect, which is typical for spatial diffusion. In the case of such a heterogeneous innovation as a nearly six-thousand-strong sample of movies, it is hard to expect unambiguous dependencies, thus the collected data lead to the confirmation of the fourth hypothesis.

The analysis of case studies in various countries makes it possible to indicate additional circumstances that had an impact on the course of the spatial diffusion. Other factors stimulating the spread of Indian cinema in various countries include top-down measures by authorities. In Pakistan these took on various forms, depending on the geopolitical situation. In times of intense conflict between India and Pakistan, Bollywood cinema presented Pakistanis as cowards and hostile antiheroes (Ud Din & Langah 2012), which led to Bollywood movies being censored or banned from Pakistani cinemas. Each time the contacts between the countries became friendlier, this was reflected in cinema – large-scale movies were made that called for peace between the countries (Kołodyński 2009). Nepal and Bangladesh had a long history of Indian movie censorship, too. These actions were caused by the countries being portrayed negatively in these movies, the desire to protect national cinematography markets, or political misunderstandings and tensions (Ud Din & Langah 2012; Khan 2012). South Africa’s ban on distributing Indian cinema in response to India’s objection to apartheid lasted for over half a century. Authorities in Australia, which was a lead shooting location for Bollywood movies, used them to present the country as tolerant and open towards foreign tourists and students (Hassam 2012). The Soviet Union actively supported the popularisation of Indian cinema as a result of long-lasting relations with India, while at the same time promoting traditional models that were alternative to those propagated by American or European cinema (Igorevna Doroshenko 2012; YanderSteene 2012). According to YanderSteene (2012), in Senegal and Nigeria the peoples using the Wolof, Fulani, or Hausa language watched movies produced in Hindi due to their short and relatively infrequent dialogue and an alleged similarity in the sounds of these languages.

It has been proven that neither the small distance of reception countries from Mumbai, nor the presence of a large Indian diaspora had an impact on the time of the premiere of Bollywood movies in the cinemas of these countries. The analysis did not show any dependence between the distance of the reception countries from Mumbai and how spatially diffused Bollywood cinema was. A relation was observed between the size of the Indian diaspora in the reception countries and the number of distributed Bollywood movies. Strong associations of cultural products, especially of Indian cinema and the shaping of national identity among the Indian diaspora, have already been proven by N. Kumar & J.-B. E. M. Steenkamp (2013), S. Rao (2010) and R. Bandyopadhyay (2008). Probably because of the neighbourhood effect, in Ireland, where the Indian diaspora is little, cinemas screened nearly 100 movies, and the first had its premiere there in only 2002. This effect can also be detected in Poland, Germany and the countries of the Arabian Peninsula. The hypotheses regarding the dependence of the start time and course of the spatial diffusion of Bollywood cinema, and the distance and diaspora size were made probable, except for the third hypothesis.

Analysis of the case studies and classification of the countries support the conclusion that the course of Bollywood cinema diffusion was significantly influenced by additional circumstances in analysing the national identity of the Indian diaspora. Including present in the selected countries. They took the form of barriers these data could create a more complete picture of the diaspora’s that inhibited or stopped the diffusion of Bollywood cinema in impact on the dissemination of its native culture, but it would be places with circumstances that facilitated the popularisation of very difficult to collect such detailed, reliable data for the more this kind of cinema (spatial and cultural proximity). These results than 70 countries analysed. The adopted timeframe allowed for may have been influenced by the nature of the tested subject of different trends to be seen in the global dissemination of popular the innovation – a very large and diverse group of movies. The Hindi cinema, but, as a result, it masked the role of ephemeral but method by which movies are accepted for cinema distribution significant trends, such as the growing popularity of Hindi cinema – a direct decision by film distributors and, in some cases, the in Europe at the beginning of the 21^st century. Subsequently, authorities – also played a decisive role. In these circumstances the deliberate inclusion of only cinematic distribution made it the expected implications were not rectilinear, although the possible to achieve the goals of the paper. However, there could results were satisfactory and reliable. be a number of factors that influenced the cinematic distribution

Finally, it is worth mentioning a few insights upon obtaining of films before they were digitised, which could have contributed the research results. Firstly, data only on the size of the Indian to other barriers that were omitted in this paper. Nevertheless, the diaspora in certain countries may seem incomplete without above considerations are proof that the problem analysed in the reference to the broader cultural, social and political context. Such above paper was complex and worth developing. a conclusion was made by, among others, R. Bandyopadhyay (2008),