Warfare tourism: an opportunity for Croatia?

Abstract This article explores only a variation of tourism – warfare tourism – in Croatia, which is a recognised tourist destination. The main objective of this multidisciplinary research is to clarify and classify young residents (undergraduate students) of Croatia into different clusters based on their views of the potential for warfare tourism development in the country. A total of 292 completed questionnaires were recorded and prepared for empirical analyses. Ward’s principal component score method, independent sample test and descriptive analysis were employed in this study. We found that two clusters of youths could be identified – believers and doubters, the latter representing the majority. The difference between clusters, in fact, is not so big: however, believers, contrary to doubters, do recognise warfare tourism opportunities for the development of war-affected areas in Croatia, opportunities that first need to be conceptualised on the strategic level (country/county/city/municipality).


Introduction
Croatia is a distinctive tourist destination of the Adriatic region and, economically, it is indeed highly dependent on tourism. One of the most significant problems is regional imbalance: tourism is most developed mainly in coastal areas, while mainland Croatia is less recognisable for its tourism (with the exception of the City of Zagreb) (Štrukelj & Šuligoj, 2014). In addition, some areas were strongly affected during the war in the 1990s, another cause of their tourism imbalance. The relevant questions here are, could warfare sites be interesting for tourists and is the homeland war interesting for tourists? What do young residents of Croatia (undergraduate students), who do not have direct war experiences, think about that? Such a local resident-, consumer-family/relatives-oriented approach is still needed in dark tourism studies where conceptual researches are still dominant (Biran, Poria, & Oren, 2011;Kidron, 2013;Seaton & Lennon, 2004;Stone, 2010;Stone & Sharpley, 2008). The aim of this multidisciplinary research is to fill this gap in the case of young Croatian residents.
Warfare tourism is only one part of the so-called dark tourism. The field of dark tourism may be considered as a micro niche of special interest tourism (Minić, 2012;Novelli, 2005).

KEYWORDS
Warfare tourism; warfare heritage; potentials; youths; croatia Foley and Lennon (1996) and Lennon and Foley (2000) have defined the phenomenon dark tourism and the concept as a research area. Dark tourism may simply be described as tourism related to visitations to sites associated with death, suffering or the seemingly macabre (Stone, 2006, p. 146), or tourism related to sites associated with death, disaster and tragedy for remembrance, education or entertainment (Foley & Lennon, 1997), or for political purposes or economic gain (Stone, 2006). This article explores only a variation of dark tourism -warfare tourism in Croatia. Whilst there are studies of tourism (development) in the country, almost none explicitly research war-related tourism in the post-war period. Furthermore, war is mentioned as an obstacle to or weakness of development; however, after two decades, poor development cannot be the outcome of those previous events, but rather the result of inadequate and non-competitive destination management (Armenski, Gomezelj, Djurdjev, Ćurčić, & Dragan, 2012).

Theoretical background
Systematic development of battlefield tourism was initiated after World War One (Hertzog, 2012;Walter, 2009a;Winter, 2009bWinter, , 2011. Today, warfare sites probably represent the largest single category of tourist attractions in the world (Henderson, 2000;Ryan, 2007;Smith, 1998; see also Weaver, 2000;Wiedenhoft Murphy, 2010). Warfare tourism is a particular form of dark tourism, and it includes visiting war memorials and war museums, 'war experiences' , battle re-enactments (battlefields tours) (Dunkley, Morgan, & Westwood, 2011, p. 860;Naef, 2013aNaef, , 2013b. Dann (1998) treated it as one of five divisions called fields of fatality: areas/land commemorating death, fear, fame or infamy. Seaton (1996) distinguishes two authentic groups directly linked to war: sites of individual or mass deaths, and memorials or internment sites; in this context Stone (2010) typifies dark conflict sites and dark camps of genocide. Henderson (2000), Siegenthaler (2002), Wight & Lennon (2007), Goulding & Domic (2009) and Stone (2010) highlight that the interpretation of war-related sites is a sensitive issue. In this context, Naef (2013a) and Baillie (2012) problematise Croatia's monopolisation of memory in the case of Vukovar. Such 'unhealthy' circumstances could provoke a new conflict in society, which could prevent the economic and tourist subsystem from developing in the right direction.
The dynamics of warfare sites visitation changes over time. At the cessation of a war, true tourist numbers begin to increase as a growing number of adventure tourists, former soldiers, their families and all others in any way linked with the war visit the sites. All these tourists are faced with still dangerous, limited and disorganised tourism products (Weaver, 2000; see also Winter, 2009b). Tourism is often a potential contributor to socioeconomic development and, regeneration, as well as a vector for integration into the global economy (Novelli, Morgan & Nibigira, 2012;Weaver, 2000;Wiedenhoft Murphy, 2010). According to Weaver (2000), (especially) large-scale wars produce large-scale surpluses of what is, paradoxically, a good war dividend 1 for socioeconomic development. Lynch (2008, 2011) use the term phoenix tourism which is less tourism centric and much friendlier to an affected community in its social reconciliation and urban regeneration. Nevertheless, warfare sites, because of their mythological status, seem less liable to the law of the product-cycle than other types of tourist attractions (Weaver, 2000). In this context it would be reasonable that war-related sites in Croatia represent one of the components of the Croatian tourism offering which is less sensitive to contemporary (global) trends and changes. Consequently, for this research, we identify the following working questions: (1) does dark tourism offer opportunities for the development of war-affected areas in Croatia, especially because of the short time distance; and (2) could local providers prepare interesting thematic programmes for tourists and this way capitalise on the war dividend?

Distinctive examples of dark tourism sites
Military history offers many interesting sites all around the world, which are also interesting for research from different aspects, including tourism (see Table 1). Many of these authentic sites have distinct conservational, educational and commemorative meaning, which must engender a degree of empathy between the visitor and the (past) victim (Henderson 2000, Hertzog, 2012Kidron, 2013;Miles, 2002;Robb, 2009). Minić (2012), Seaton & Lennon (2004), Wight & Lennon (2007), Walter (2009), Biran et al. (2011, Lee, Bendle, Yoon & Kim (2012), Ozer, Ersoy, and Tuzunkan (2012) and Stone (2012), in this sense, argue that dark tourism is only the culture's subtype and part of heritage or its special form of expression.
The possible bondage between tourism and war memories was recognised by Slade (2003) and Winter (2009b), who put forward the battlefield of Gallipoli (Turkey), which has de facto psychological and cultural origins and a strong nationalistic hint. Moreover, this destination has important implications for the construction of nationhood and notions of mythmaking for Australians and New Zealanders, where many of them are not interested in death itself (Slade, 2003) or for systemic and unrelenting militarisation of Australian history and culture, especially in relation to the youth (McKay, 2013). Hence, warfare destinations could also create negative/wrong international imaginary equating the place with arms and bloodshed; Wight and Lennon (2007) cite Lithuania as a destination with a dark heritage of invasion, genocide, and repression. If we turn to the World War Two-related sites across the world, Siegenthaler (2002) in his research of tourist guidebook presentations of Hiroshima and Nagasaki discovered the denial of the war's memory, meaning the containment of national war memories, ambivalence and the decentring of historical perspective. In addition, only a small share of young Nagasaki visitors sees this site as an attraction. It is noticeable that young people are seeking more comprehensive information about the past and want to understand war atrocities (Cooper, 2006). In this context, Thurnell-Read (2009) -in the case of Auschwitz-Birkenau -suggests two possible measures relevant for  Gilbert, 1986;miles, 2002;ashworth, 2002;jilovsky, 2008;thurnell-Read, 2009;Biran et al., 2011;cohen, 2011;kidron, 2013 vietnam  the youth: (1) achieving a greater understanding of the historical facts of the Holocaust; and (2) the affirmation of humanist values as understood, at times ambivalently, with reference to contemporary society. It is also significant that Auschwitz-Birkenau hosts a heritage experience rather than a merely dark tourism one (Biran et al., 2011). Lennon and Foley (2000) claim that sites/events which are more recent are darker than those which have a longer history, although Causevic and Lynch (2011), in the case of the memory of the battle of Kosovo, refute this claim. The authors, according to their research in Bosnia and Herzegovina, claim that memorabilia of war are core to the concept of post-conflict tourism development, where using the dark tourism context reduces their meaning to a rather narrow tourism context; in this meaning dark tourism should be supplemented by long-term frames and indigenous perspectives and not only oriented to Western hegemonic tourism constructions (Lee et al., 2012;Robb, 2009). Another recent example of a destination with great potential for dark tourism is Croatia with some specific practices. Here we can see that national culture, memory and economy are mutually embedded (Rivera, 2008), which means that the past, history and heritage are closely aligned to the dominant political system and, as such, the past can also be used as a vehicle to promote political views and Croatian 'nationhood' (Goulding & Domic, 2009, 99). In these circumstances, Vukovar has become a Croatian mythical place where the visitors can see the effects and remains of the homeland war, where younger generations can not only understand the experience of war but also feel the suffering of the Croatian people, and listen to the story of the rise of independent Croatia (Kardov in Naef, 2013a). Contrary to Dubrovnik, Vukovar was never an overly popular tourist destination, but it appears that its symbolic status has attracted a number of visitors. The tourism which accompanies the creation of memorials is often seen as 'nationalistic' tourism intrinsically linked to the symbolic status of the city (Baillie, 2012;Naef, 2013a). Regardless of its status, on the strategic level, Vukovar-Syrmia County has not planned the development of dark (warfare) tourism (Razvojna strategija Vukovarsko-Srijemske županije 2011-2013, 2011), 2 although this should be its flagship product. Moreover, a brief review of strategic documents 3 on the national level (see Croatian Tourism Development Strategy till 2020, 2013) shows that Croatia has no strategic plans on this subject. We reached the same result upon reviewing previous or current strategic documents of some war affected counties in the 1990s. 4 Zadarska County (Glavni plan razvoja turizma Zadarske županije 2013 -2023, 2013) has exceptionally recorded all warfare examples as heritage (mainly from World War Two), but there are no strategic objectives related to their inclusion in the tourism offering. Consequently, the relevant question is, what do young residents, who study tourism or other business sciences, who do not participate in decision-making and who do not have direct war experiences, but who have listened to the stories, 5 think about that? Do they have similar views? These dilemmas have not been resolved in past research either in Croatia or abroad, although there is much research where the war of the 1990s is discussed from different points of view. The questions listed, together with the questions raised in the previous section, aimed to design and empirically verify the following research assumption: Young residents (undergraduate students) of Croatia mostly support warfare tourism development and can be divided, on the basis of their views on the potentials of the warfare tourism, into two or more statistically significantly different groups.

Methodology
Data for this study were collected from the undergraduate students of tourism -and/or business-oriented university departments, faculties or universities of applied sciences in Croatia. Students belong to a generation that has no direct experience of the war. In the country, eight public tourism-and/or business-oriented university departments and faculties were invited to participate in the study, and seven responded to the invitation. Using a Monte Carlo sampling approach, we came to six public universities of applied sciences, inviting them to participate in the study, and four (Požega, Knin, Karlovac, Čakovec) responded to the invitation. Before collecting the data, questionnaires were tested in two phases: (1) (technical) testing on a sample of 500 automatic computer-completed questionnaires; and (2) testing and discussing with 10 Croatian postgraduate students (mostly PhD students). After a few corrections, undergraduate students of the participating institutions were then asked to fill out the web surveys. Students could fill out the survey in the classroom or in peace at home. In total, 361 at least partially or 292 fully completed questionnaires were recorded 6 . The survey was conducted in spring 2014 and lasted 45 days.
Respondents were mostly of Croatian nationality (92%), 2.8% were of Serbian nationality, and 2.1% were Bosniaks. The others represent less than 3% all together. Among the respondents, 69.4% of the sample were female and 30.6%, male. They come from all Croatian counties, although the largest proportion comes from the Osječko-Baranjska County (14.8%). The largest share of students were 19 years old (26.8%), followed by 21-year-old (25.1%) and 22-year-old (17.2%) students; all other groups represent shares of less than 15% each.
For the study, we used a questionnaire with 68 variables in the Croatian language, but only 14 were relevant for this research. Besides demographical items, there were also items with a five-point Likert scale, ranging from 1, 'strongly disagree' , to 5, 'definitely agree' , or 1, 'not at all probable' , to 5, 'completely probable' . The relative part of the questionnaire included a series of items addressing students' perception of the importance of warfare sites/events visitation. This was followed by a set of statements designed to clarify their understanding of homeland war sites/events, dark tourism potential, and visitation in the future. The questionnaire was based on some previous research (i.e., Biran et al., 2011;Kim, 2009;Stone, 2010) and adapted to Croatian circumstances. SPSS 20.0 software was chosen as the mechanism to collate survey data, as well as to analyse and present the results.

Results
An empirical analysis was conducted in three phases. The first phase involved undertaking a cluster analysis of the selected items to see whether there is a group with a strong approval of warfare tourism in Croatia. The second phase focused on results checking, using additional statistical methods, and the identification of statistically significant differences between the groups. The purpose of the third phase was to identify and describe the respondents classified into different groups.
Cluster analysis is a technique that almost invariably produces two or more groups that are significantly different from each other. Ward's principal component score method with Euclidean Squared distance metric was employed in this study to form the groups of young residents (undergraduate students) of Croatia. A total of 292 young respondents expressed their opinion on the potential of warfare tourism in Croatia, which includes four variables: Q1a: Visiting significant buildings, monuments, museums, etc. and attending events related to the war in Croatia in the 1990s, I understand to be an example of a special form of tourism, so-called dark tourism 7 ; Q1b: With regard to the war of the 1990s, Croatia has the potential to become a recognisable destination of dark tourism; Q1c: Croatia should develop special tourism packages and tours for organised visitation of significant buildings, monuments, museums, etc. that are related to the war in Croatia in the 1990s; Q1d: Areas in Croatia, which were heavily damaged during the war in the 1990s could now develop better and faster with the help of dark tourism.
First, we calculated Cronbach's α to test for reliability. For all four variables, the coefficient was 0.808, showing a strongly reliable set of variables. Clusters can be identified by analysis of agglomeration schedule and dendogram. In our case, the first phase revealed that two clusters of young respondents could be identified -see ( Figure A1 in the Appendix). Clusters will be described in the following paragraphs, as we first have to check the statistical significance of difference between clusters (second phase). We carried out the t-test method for independent samples to determine the difference in youths' views in different clusters. T-test confirmed a statistically significant difference between the mean values of C1 and C2 when p = 0.05 -see Table 2. That implies that there is sufficient evidence to conclude that youths' views in different clusters are different regarding all analysed variables. In order to form more specifically defined clusters, some additional variables and methods were used as well, all of which are described in the following text.
Although variables used for carrying out cluster analyses are substantially similar and close to each other, 8 the most obvious differences emerged regarding Croatia's potential of becoming a recognisable destination of dark tourism (Q1b). Members of C2 here clearly had the most concerns. Moreover, members of C2 cannot identify warfare sites as dark tourism sites (Q1a) -see also Table 3, what is a crucial precondition for future dark tourism development.
According to these basic findings, in the next phase we defined and described both clusters derived from the cluster analysis. The clusters identified are:

Cluster 1 (C1): believers (supporters)
They form 46.6% of the entire sample and score highly on supporting the idea that Croatia is a destination with potential for dark tourism product development and economic recovery of the affected areas. High average ratings for all items are shown in Table 2. Hence, data in Table 3 show a distinct asymmetrical distribution of ratings where cluster members mostly agree with the items (Mo = 4) and only a small proportion of them are neutral or do not agree. We also asked respondents whether certain significant buildings, monuments, museums, etc., related to the war in Croatia in the 1990s should be made available for visitation (Q5) and whether attending certain significant events related to the war in Croatia in 1990s (e.g., commemorations, anniversaries, etc.) is important (Q6). Members of the cluster believe that some significant physical sites (Q5) should be available for visitation (x = 3.70; Mo = 4). In addition, members also think that attending certain significant events (Q6) is important (x = 3.61; Mo = 4). These are additional variables included into the research in order to better describe the cluster. The descriptive method was employed to reach this aim.
In the next step of the research, we focused on the respondents' future actions regarding dark tourism sites in Croatia, which means that four more variables were introduced. We asked them about the probability of their future visits to buildings, monuments, museums, etc. that are related to (any) war in Croatia (3a); about the probability for their future visits to buildings, monuments, museums, etc. that are related to the war in Croatia in 1990s (3b); and about the probability of recommending visits to buildings, monuments, museums, etc. and events related to the war in Croatia in the 1990s to Croatian residents (3c) and to foreigners who intend to come to Croatia (3d). First, Cronbach's α is 0.916, indicating a high level of internal consistency for our scale with this specific sample. Second, independent samples t-test made by considering the results of cluster analysis shows that members of C1 significantly differ from the members of the other cluster (Table 4). Comparing data to Table 2, we see here that results are more centrally oriented for both clusters. However, members of C1 are slightly more confident that they will visit warfare sites or events in the future. Furthermore, they will also recommend visits to others, domestic visitors and foreigners. Personal recommendations and word of mouth are traditionally the most effective promotion in tourism, and this is extremely important for dark tourism development in Croatia as well.  According to all listed results and explanations, we named C1 members believers or supporters. They are characterised by the belief that dark tourism could be a significant part of Croatian tourism with favourable effects on the economic development of war-affected areas. In addition, they are slightly in favour of warfare sites and events visitation and these they would recommend to others as well. We would also like to highlight that, in our research, 66.2% of believers are female, what is close to the gender distribution of the entire sample. We can claim the same for national distribution.

Cluster 2 (C2): doubters (waverers)
This cluster is not significantly larger than the first one and includes 53.4% of respondents, of which 69.2% are female. National diversity is mainly formed by Croats (88.5%), Serbs (3.8%) and Bosniaks (2.5%). As we mentioned earlier, C2 members do not identify the potentials of warfare tourism for Croatia ( Table 2), but we have to take into account that they are not categorically against this type of tourism (mean values tend toward the middle of the scale). Data in Table 3 confirm a symmetrical central distribution of ratings where cluster members are mostly neutral (Mo = 3), with slightly more of those who do not agree/ do not agree completely compared with those who agree/agree completely. In this context we cannot call them opponents but rather doubters or wavers. The views related to Q5 and Q6 where C2 results were practically identical to those of believers (Q5: x = 3.68; Mo = 4 and Q6: x = 3.61; Mo = 4) were also very interesting. These additional variables confirm that C2 members could also support warfare sites/events visitation. The inconsistency observed when comparing the data in Tables 2 and 3 could be understood in the light of their general neutral position, where some single variables seem to be more acceptable to respondents, especially when they are not treated as dark tourism sites/events. This claim is in line with the variable 1e (Table 2), which has the lowest value.
Variables in the Table 4 are all centrally oriented, which means that members will perhaps visit warfare sites/events related to (any) war in Croatia or war in Croatia in the 1990s and that they will perhaps recommend visits to these sites/events to Croatian residents and foreigners who intend to come to Croatia. However, this indecision or uncertainty definitely separates them from the C1 members.

Discussion and conclusion
This article has explored some of the key challenges associated with war-related tourism development from the perspective of young people (undergraduate students), taking Croatia as being illustrative (although not necessarily typical) of the many specifics in a post-conflict country which aspires to harness its tourism potential to aid recognisability and economic growth. Based on the findings of this research, we claim that one part of Croatia's young population (undergraduate students) believes in warfare tourism potential, while the rest are not absolutely against it, meaning that perhaps in the future, through systematic awareness, they would be able to identify opportunities, especially in war-affected areas. As Štrukelj and Šuligoj (2014) point out, cultural heritage sites provide an excellent base for development also to less-developed states and regions that lack other natural resources and industry. This link with culture is largely evident in the case of Croatia, where national culture, memory and economy are mutually embedded (Rivera, 2008), which should foster warfare tourism development, if it is not abused for political purposes. The results of our research indicate that visits to warfare sites/events become problematic when they are placed in the context of tourism. This does not support the assertion of Novelli et al. (2012), Wiedenhoft Murphy (2010) and Weaver (2000) that tourism is often a potential contributor to socioeconomic development and regeneration and a vector for sites/events integration into the global economy.
The main research assumption defined in previous sections could only be partly confirmed. The results of a cluster-analytic approach demonstrate that believers statistically significantly differ from doubters who represent the majority. Results are representative for the involved population.
So as not to remain only on statistical construct and theory, we described and identified clusters with different characteristics. Here we found that the difference between clusters, in fact, is not so great. Believers, unlike doubters, do recognise warfare tourism opportunities for the development of war-affected areas in Croatia. Hence, they think that interesting thematic programmes for visitors/tourists should be designed. This is extremely important if regional/local providers want to capitalise on war dividends (see Weaver [2000] or Antolović and Škare [2006] in relation to monument annuity), but such programmes should first be conceptualised on the strategic level (national/regional). The potential practical applicability of the proposed research should be underscored from a destination planning and marketing perspective. It allows destinations to be consciously aware of warfare tourism as a vehicle for their development, and to plan and market their offerings based on the various attractions. As a result, these findings can provide a meaningful starting point for educational institutions in tourism and business to consider how to effectively design and provide their curriculums without neglecting a certain part of the past.
We should take into account the claim of Lennon and Foley (2000) that sites/events which are more recent are darker than those which have a longer history, as well as the claim that areas of recent conflicts still have open wounds, memories and emotions which may disable normal regional/local socio-economic development (Causevic & Lynch, 2008. The impact of these elements on believers and especially on doubters was not researched. In addition, researchers could also investigate the current socioeconomic situation in relation to the homeland war and tourism in other (war-affected) counties. In any case, warfare tourism remains an under-researched topic in South-eastern Europe and yet, in addition to holding much potential, it is also less susceptible to any global downturn. This article seeks to fill this void, because it is the first to deal with warfare tourism in Croatia in relation to a generation that did not participate in the homeland war and will be the main developer and operator in Croatian tourism in the future. Notes http://bpz.hr/_Data/), Šibensko-Kninska (Retrieved April 1, 2014, from http://sibenskokninska-zupanija.hr/upload/stranice/2013/07/2013-07-18/67/dokumenti/Razvojna_ strategija.pdf; Održivi razvoj u ratom stradalim područjima Hrvatske, 2005), Karlovačka County and Ličko-Senjska County (Bro, 2008; Retrieved April 1, 2014, from http://www. europski-fondovi.eu/sites/default/files/dokumenti/). 5. Veterans and survivors will remember the war in a different way than, say, their children, or unrelated individuals for whom it has become history (Stone 2012;Walter, 2009). 6. Sample represents approximately 0.6% of all students of economic and interdisciplinary social sciences with a higher female-to-male student population. 7. We did not observe an adequate expression in the Croatian language for warfare tourism. 8. Spearman's correlation coefficient indicates statistically significant dependence between all four variables. The calculated values range from 0.488 to 0.573 at the 0.01 level (2-tailed).

Disclosure statement
No potential conflict of interests was reported by the author.