1. Introduction

1 The aim of this article is twofold: to propose a more specific definition of some terms used in the context of interviews with photographs, and to present an interview technique, which is still little known.

2 About ten years ago Clark-Ibáñez (2004) wrote that the Sociological Abstracts database contained 140 studies, which mentioned the keyword ‘photo’ in their abstract, 80 of which used photographs as an integral part of the study. To confirm the importance that images have acquired in social research, the Sociological Abstracts database contain today (January 2014) more than a thousand publications that include the keyword photo in their abstract. Given this high number, I was not able to read all these documents to examine the specific ways in which the photos were used but I have noted, in the field of the studies belonging to the ‘sociology with images’ (Harper, 2000 ; La Rocca, 2007), that the expressions photo interview or photo elicitation are used sometime with almost no distinction. For this reason, in the next paragraphs I try to point out some differences inherent in these interview techniques and to propose a more systematic use of certain terms. This choice was also made to introduce another section in which I show the technique of bipolar photo elicitation that is still little known and that, until now, has only been used in the two studies which are presented here. I conclude the article with some methodological reflections on three strengths of this technique.

2. A definition of the area of photo elicitation and an overview of this one

3 In interviews conducted using photo images it is possible, first of all, to make a distinction concerning the photos themselves, which can basically be divided into two types:

4 a) photos that originate from the researcher, who shot them based on his/her interpretation of the reality he/she is investigating and that he/she will share with the interviewee in order to find other meanings. Alternatively, the researcher can select already existing images that will be used with the same criterion;

5 b) photos that originate from the interviewees, who are invited by the researcher to shoot some photos that represent the reality in which they live or a specific sector of that reality. The interviewees too can, alternatively and depending on the type of research in question, present already existing images.

6 It is possible therefore to affirm that the main differences between the two types concern, firstly, the participation or non-participation of the interviewee in the shooting of the pictures and/or in their collection and, secondly, the fact that the photos are shot from the etic point of view of the researcher or from the emic one of the interviewee. The term emic identifies in fact the categories of thought of the native subject, his/her view of the world, the local/popular notions and concepts shared by his/her culture or by the human group that is being studied. The term etic refers to the constructs, descriptions and analysis formulated according to the conceptual terms established by the scientific community (Lett, 1990; Olivier de Sardan, 1998).

7 The scientific literature is ripe with more accurate definitions for the research techniques that confer the task of taking pictures by the interviewee. As a starting point of Hurworth (2003) to take some examples, we can mention the autodriving technique, initially used in the marketing sector by Heisley and Levy (1991) who, with the aim of increasing the involvement of the consumers interviewed, asked them to shoot pictures to be commented upon in a second encounter. The term autodriving was used by the authors because « the interview is ‘driven’ by informants who are seeing their own behaviour » (Heisley and Levy, 1991, p. 261). However, in the title of their study, the authors describe autodriving as A Photoelicitation Technique.

8 With the reflexive photography technique (e.g. Douglas, 1998; Berman et al., 2001), the participants are given a camera and are asked to shoot photos that represent their impressions on a specific subject or photos of people, places and events. The participants’ reflections on the photos are later explored in the interviews.

9 Wang and Burris (1997) proposed the photovoice technique, described as a process through which people can identify, represent and improve their community by means of the photographic technique. This technique, too, involves at a later stage the discussion of the photos which, according to Wang et al. (2000), can be used for participatory action research and as a means of enabling personal and community change.

10 The techniques are frequently described as involving two phases : picture taking and discussion with the researcher. Only few authors make a terminological distinction between the two phases: Oliffe and Bottorff (2007), for example, use the photovoice technique to acknowledge participants as the authors of the photographs and they add the term photo elicitation to describe the process by which the photographs are discussed during individual interviews. van den Beemt et al. (2010), too, use the autodriving visual elicitation and then the photo elicitation; Schulze (2007) uses reflexive photography first and photo elicitation in the last phase. We must remember that, in these cases, we are dealing with images shot and/or collected by the interviewees.

11 To avoid potential terminology confusion, I propose here a distinction that, in my opinion, could contribute to a clearer definition of these visual techniques. Kolb (2008) uses the term participatory photo interview to describe her research technique that consists of four phases, including both the shooting of photos by the participants and the following discussion thereon. However, the title she chose for the relevant paragraph is Phases of the photo interview. Jorgenson and Sullivan (2010), too, use the expression participatory photo interview and describe the procedure as composed of the autodriven and the photo elicitation phases.

12 My proposal is to use participatory photo interview as term inclu­ding the research techniques that involve both the shooting and/or the collection of images by the interviewees (that can further be defined as autodriving, photovoice and so on) and the subsequent discussion on the photos with the researcher, for which we will use the expression photo interview.

13 This will allow us to use the term photo elicitation exclusively for the type of interview in which it is the researcher who shoots and/or chooses the images to be shown to the interviewee and on which to base the discussion. Through this procedure, the interviewees are actually elicited by the visual stimuli proposed by the researcher that, most probably, they have never seen before.

14 The following description, in part, is valid for the photo interview too, but I refer here, in particular, to the photo elicitation as just described.

15 « Photo elicitation is based on the simple idea of inserting a photograph into a research interview » (Harper, 2002, p. 13). The photographs presented by the researcher are used as a basis for the interview and play a specific role in it, as they help the interviewees recall memories - or to reflect on various issues - that, without images, would not come to their minds (Banks, 2001). This happens because

16

the parts of the brain that process visual information are evolutionarily older than the parts that process verbal information. Thus images evoke deeper elements of human consciousness than do words; exchanges based on words alone utilize less of the brain’s capacity than do exchanges in which the brain is processing images as well as words. These may be some of the reasons the photo elicitation interview seems like not simply an interview process that elicits more information, but rather one that evokes a different kind of information (Harper, 2002, p. 13).

17 Using this technique, the researcher and the interviewee can discuss the images that represent the reality (or a part of it) where the interviewee lives and/or that he/she has experienced. Through the photos, the interviewee’s narration can start from the concrete, cataloguing and describing the objects that appear in the picture, to then move to abstract thoughts, explaining the meanings that these objects have for him/her (Wagner, 1979; Collier and Collier, 1986; Harper, 1988; Harper 1994; Harper 2002; Faccioli and Losacco, 2003).

18 The meanings enclosed in the researcher’s questions, thus connected to his/her etic concepts, may not include the interviewee’s categories that are located on a different level (Pretto, 2011). But if the interviewee concentrates mostly on the image, he or she will not be influenced by a question posed by the researcher on the basis of his/her conceptual categories only. Using photographs, the researcher does not have to name or define an object and in this way, he/she avoids giving it a positive or negative connotation, leaving the interviewee free to interpret it based upon his/her personal life (Harper, 1994). The photo can stimulate the interviewee to better explain his/her « world of meanings » and to realize that « the researcher may be very interested in what he/she takes for granted » (Faccioli and Losacco, 2003, p. 35).

19 Often, the power of the image in an interview setting lies in the ‘weakness’ of the stimulus showed and in its polysemy, which gives anyone the chance to read it from a personal perspective and, on this basis, to interpret it and give it meanings that could be already present in the mind (Becker, 1986; Ferrarotti, 1992; Harper, 2002). The strength of this approach lies in the ‘talk back’ of the images to the interviewee, that allow him to consider them from different points of view (Rose, 2001).

20 Oliffe and Bottorff (2007) underline that the use of images in interviews can benefit the researcher in three ways. Firstly, the use of images can help the researcher to overcome the natural barriers that the interviewee may erect when he/she has to answer direct questions that could « put control, autonomy, or rationality into doubt, if only implicitly » (Schwalbe and Wolkomir, 2001, p. 91): the use of photos activates a ‘show-and-tell’ mechanism that facilitates the flow of speech in the interviewee. Secondly, given the fact that the interview can point to very private and highly emotional issues, the participants may find relief and shelter in pictures, for example by talking of an image in the third person or referring to other people. In this sense, pictures offer the interviewee a strategy to take some distance from the personal dimension and/or from painful experiences and this facilitates self-disclosure while protecting the interviewee’s anonymity. In my opinion, these two aspects are particularly useful when the object of the research may be ‘delicate’ for respondents (as in the two examples in the next section). Thirdly, photos require a ‘thick description’ (Geertz, 1973) with details and explanations that contribute to reduce any misunderstandings on the part of the researcher. The latter becomes a sort of spectator, losing his role of interviewer, while the interviewee takes on a teaching role (Flick, 2002).

3. The bipolar photo elicitation: two case studies

21 A particular type of photo elicitation, used for the first time by Faccioli and Zuccheri in 1998, was recently applied by Pretto and Battello (2014) [1]. The special feature of this technique is that a set of two images – and not one single image - is shown to the interviewee. To present this type of interview I am here going to describe the way in which this technique was applied in two case studies, which are also the only two of which I am aware.

3.1 Case 1: the ambivalent perception of alcohol

22 The first case study concerns a research study commissioned by the municipality of Bologna with the aim of preparing an information and education campaign on the risks of alcohol use and abuse. The study investigated both the perception of the health risks connected to alcohol consumption and the existence of taken- for-granted cultural attitudes for which alcohol use (culturally legitimated and therefore ‘good’) and abuse (culturally disapproved and therefore ‘bad’) are considered two distinct and unrelated behaviours. In other words, if we imagine alcohol consumption as a continuum whose ends are use and abuse, it is difficult to decide where to draw the line between the two. And without a clear separation, it is easy to cross the line without being aware of it. This risk is even more serious if we consider that alcohol, in the Italian culture, in the form of wine, is considered as food and is almost always on the table at meals. Moreover, alcohol is present when there is something to celebrate and to mark happy events and rites of passage. Non-drinkers, teetotalers for example, are sometimes distrusted (Faccioli and Losacco, 2003).

23 On the assumption that alcohol use and abuse are two faces of the same coin, the success or failure of the information campaign depended on the fact that the public to which it was directed could identify or not with the above mentioned continuum. If people think that their alcohol consumption is ‘good’ and that it is totally disconnected from the ‘bad’ consumption, they will think that the campaign does not concern them. Only the awareness of the continuum will make them sensitive to the message.

24 Faccioli and Zuccheri (1998) have collected 400 open-ended interviews and have also created and used a tool that can detect the interviewees’ awareness (or unawareness) of this continuum or, in other words, the ambivalent perception of alcohol: good and bad, legal and illegal, normal and abnormal, approved and disapproved. This ambivalence paved the way to the idea of bipolar photo elicitation based on coupled images. In each set, the two photos depict the two extremes of the continuum for the phenomenon at issue. The situations depicted in the images selected by the researchers encouraged the interviewees to give information on their idea of normality and abnormality in terms of alcohol consumption and more or less legitimate consumption circumstances. This however did not exclude the possibility for the interviewees to place themselves in an intermediate position along the continuum and not at its ends.

25 The researchers chose the images after having examined a large number of photos that appeared in widely circulated magazines, selecting those clearly depicting alcohol consumption. I provide here as an example one of the five sets of photos used in the interviews [2]:

Photo 1: Lunch (or dinner) with friends

Photo 1: Lunch (or dinner) with friends

Photo 2: Domestic violence

Photo 2: Domestic violence

26In photos 1 and 2 alcohol appears in a moment of everyday life (a lunch or a dinner with friends) and in a representation of domestic violence. Presenting each set of images, the researchers asked the interviewees:

27

to describe the two pictures separately and to talk about the meaning of alcohol in each of the situations. To compare and discuss the similarities or differences between the two photographs (this was helpful to elicit their judgement in terms of values and beliefs towards the use/abuse of alcohol). And to create a “narrative relationship” between the two pictures; for example to imagine whether people represented in the first photo might ever find themselves in the situation described in the other photo. We asked our subject to imagine how and why this could be possible (this was helpful in eliciting their imagination towards the possibility of changes related to the use/abuse of alcohol) (Faccioli and Zuccheri, 1998, p. 77).

28 In the set of photos 1 and 2, for example, the narrative relationship could be that one of the lunch participants, having drunk too much in a socially accepted setting, was violent with his wife once back at home.

29 On a methodological level, the authors observed that the answers to open-ended interviews were in general shorter than those inspired by the photos. Also, the answers in the non-visual part of the interview were more rationalized, while those inspired by the pictures were more stereotypical. The authors point out that, in general, the bipolar photo elicitation helped them to discover far more about peoples’ habits and personal relationships with alcohol than they expected.

30 In the article of Faccioli and Zuccheri (1998), as well as in the subsequent book (Faccioli and Losacco, 2003), there are no additional methodological reflections on this technique besides these ones. However, I decided to describe this case study not only to pay tribute to the creators of this type of interview but also because I used precisely these two photos to test this tool. Also, these photos help me to explain the strengths of this technique in the final methodological reflections.

3.2 Case 2: The sense of belonging in a cross border area

31 The case study conducted by Pretto and Battello (2014) concerns, too, a sort of continuum, but in this case it is a territorial continuum. The study was developed in two cities: Gorizia, an Italian city in the region of Friuli-Venezia Giulia, and Nova Gorica, in the Slovenian region of Goriska. The particularity of these two cities resides in the fact that they used to belong to the same nation, whereas now they are separated by a national border.

32 In order to give a brief explanation, Gorizia was part of the Habsburg Empire from the beginning of the 16th century and, in 1921, after World War I, it was annexed to Italy. During the Italian fascist regime there were violent ethnic conflicts in Gorizia and the Slovenian minority, which had been in the city for centuries, was discriminated against in many ways including with violence. At the end of World War II, the Germans left the city and Tito’s Yugoslav army occupied it but, in 1945, Gorizia was assigned to Italy again. The Paris peace treaty in 1947 defined the borderline between Italy and Yugoslavia: the new border divided the provincial territory but also the city of Gorizia. The Socialist Federal Republic of Yugoslavia was assigned the northern and eastern part of the suburbs and almost two thirds of the province, and a number of public buildings and facilities, like the Montesanto railway station in Gorizia, on the Transalpina rail route. The historical square in front of the Gorizia Montesanto station was divided in two by a metal fence that, for decades, was the symbol of an unnatural border which, all of a sudden, divided many families from their relatives and even many houses from their barns and fields. In order to establish a local political and administrative entity to govern this vast area, the Yugoslav regime built from the ground up the city of Nova Gorica in the urbanised part under its control (Angelillo et al., 1994). Therefore a new Gorizia was built very close to the old one.

33 Until the accession of Slovenia to the European Union, in 2004, it was impossible to cross the Italian-Slovenian border without a passport (or a special pass for the locals). However, part of the Slovenian population remained in Gorizia and in 2001 was acknow­ledged as a linguistic minority by the region of Friuli-Venezia Giulia.

34 The aim of our study was to explore the development of socio-territorial belonging in these two cities which once used to be a single multicultural reality within the same State. Since their separation, their history and social processes took different paths, but the two cities are located along the same border and the feeling of belonging to one or the other may have developed in a similar way. From the classic sociological issues to the most recent themes, our research questions focused on the implications of affective and traditional ties (e.g. Pollini, 2005) and the attachment to the territory (e.g. Pareto, 1916; Weber, 1922) on the feeling of belonging ; on the symbols of the identity of a group, the symbols that for ‘the others’ identify a determined group (e.g. Durkheim, 1912); on the sense of belonging in the Gorizia and Nova Gorica residents - to see if they feel they belong to their respective cities on the opposite sides of the border or if they accept the cultural differences and move easily from one side of the border to the other (Pollini, 1987; Pollini, 2005; Roudometof, 2005; Gustafson, 2009; Pichler, 2009).

35 I am not here going to give details on the study itself because, in this paper, I will focus on the reasons why we chose the bipolar photo elicitation and the way we used it. The first reason why Battello and I chose this technique is that the study was to take place in a dual setting: the interviews in fact were conducted both in Gorizia and Nova Gorica and involved both Italians and Slovenians. In total we interviewed 43 people: 20 Slovenians resident in Nova Gorica and 23 Italians living in Gorizia (5 of whom belonging to the Slovenian minority, a category of people that we had not considered at the beginning of the research). The interviewees lived in different States and belonged to different cultures so, for each set of photos that we presented to them, one photo depicted Gorizia and the other Nova Gorica. In some cases, the photos showed border places or areas that the two cities have in common.

36 The second reason is connected to the tormented history of these two cities and its painful consequences (and hard feelings in some cases) on the members of the two communities. By presenting images of the two cities simultaneously, we hoped to express a sense of neutrality, in order to let the interviewees feel at ease no matter their nationality.

37 The third reason, finally, is connected to potential linguistic difficulties. We personally did not speak Slovenian and were not able to find an experienced interviewer who spoke it, and we hoped to overcome these linguistic difficulties thanks to the Slovenian residents of Nova Gorica [3] who spoke Italian. Using images before asking the questions was a way to facilitate communication and so it was.

38 We have used five sets of pictures: we selected eight photos among the many that we shot personally in the two cities. After having looked up many books and Internet sites, we have selected two more old pictures of places that are totally different today. Each set was shown to the interviewees instead of the opening question. The researcher asked the interviewee this: «Could you please talk about these photos? You can start from any of them, as you prefer…». This allowed the researchers to avoid any connotations of the images. Verbal stimuli in order to obtain a more in depth analysis of some issues were suggested, when necessary, only based on the spontaneous descriptions and narrations of the interviewees. Here are, with a brief description, two of the five sets of images used.

Photo 3: Central area of Nova Gorica

Photo 3: Central area of Nova Gorica

Photo 4: Central area of Gorizia

Photo 4: Central area of Gorizia

39Neither of the cities has a true ‘centre’ because of the division of the territory; picture 3 depicts one of the streets considered the centre of Nova Gorica, while photo 4 is a picture of Piazza della Vittoria, in Gorizia, that many residents consider the city centre. This first set aimed voluntarily at giving a generic representation to introduce the theme of the two cities. Since urban landscapes are an expression of identity and, at the same time, shape the identity of the people who live around them (e.g. Hall, 2006), we decided to use this set of images to open the interview, in order to verify if the interviewees recognise and visit only the places of their city or also of the nearby one. The knowledge of the two cities expressed by the interviewees could help us to understand if and to what extent their socio-territorial belonging concerned only one of the two cities (belonging to the local dimension) or if a number of interviewees (or all of them) also felt (for some reason) bound to the other city as well. That is why the second set of photos (not presented here) concerned less recognizable areas of the two cities. The third set of pictures (not presented here) showed the Gorizia Castle and the Nova Gorica City Hall; the aim was to depict the most ancient buildings of the two cities, that can be considered as their symbols (e.g. Johnson, 2002).

Photo 5: Plaque at the current border

Photo 5: Plaque at the current border

40 These photos concerned the border and therefore the areas that the cities have in common; photo 5 is very recent (the researchers shot it in 2013) and depicts the plaque that was installed on the square in front of the Montesanto railway station in 2004 to celebrate the opening of the border with the access of Slovenia to the European Union. Photo 6 was shot in 1947, when crossing the border from Gorizia to Nova Gorica was very difficult: it portrays the Casa Rossa - Rozna Dolina border crossing, named after the first Slovenian village you encounter on the Slovenian side if you entered the country from that point of the city. We thought that this was an easily recognisable image, albeit old, because the Casa Rossa border crossing was one of the few places where Italians and Slovenians could cross the border with their passport or with the special pass (prepustnica) for ‘border residents’, which could not be used more than four times a month. Crossing the border today is very simple: just walk over the plaque you see in photo 5. In this case we chose one photo shot by the researchers and an old one, to enhance the changes that took place in the border area. Photo 6 also shows a sign saying ‘Yugoslavia’, so the image dates back to a time when Slovenia was not yet an independent State. With this set of images we wanted to explore how the border is perceived today in Gorizia and Nova Gorica: now that there are no physical barriers would there still be emotional barriers?

Photo 6: Casa Rossa - Rozna Dolina border crossing

Photo 6: Casa Rossa - Rozna Dolina border crossing

41 The fifth set of photos, finally, represented the road signs currently in use at the border, which include the symbols of the European Union.

4. Methodological reflections

42 In general, the photo elicitation is considered as a technique very suitable for the study of experiences and is often used when the researcher is interested in the individuals’ interpretations and/or perceptions of specific circumstances or events (Carlsson, 2001). So, in my opinion, the bipolar photo elicitation is a valuable tool for studying those that Bertaux (1998) defines as ‘specific areas’, i.e. areas of study that touch on themes that could be very ‘delicate’ for the interviewees. In particular, there are three strengths that I found in this technique compared to the photo elicitation with one image.

43 First: to reduce the embarrassment of the interviewee. Thanks to my previous research (Pretto, 2006; Pretto et al., 2012) I am still in touch with some ex-alcoholics who agreed to observe and comment on the photos 1 and 2 used in this article to describe the case study of Faccioli and Zuccheri. In order to test the tool I showed them the set of photos and I asked them to talk me about them. Not only that: I also asked them to say what they would have thought if I had showed the photos one at a time. As I explain later, I repeated the same test with people who had no problems with alcohol.

44 The ex-alcoholics answered in the same way: seeing only the photo on domestic violence, they would have felt a great sense of embarrassment and shame. This, not because each of them have beaten their wives when they were drinking (indeed, two alcoholics were women), but because the image reminded them of the various moments in which they lost the control of themselves due to the alcohol. So, showing that single photo to interviewees in (possible) trouble with alcohol, I could have embarrassed them. Instead - they told me – looking at both the photos they were able to speak about moments in which alcohol is used to celebrate with friends or at happy events, reducing the embarrassment.

45 Battello and I chose only this technique to carry out our research in Gorizia and Nova Gorica just thinking that the inhabitants of the two cities shared a ‘delicate’ past. Also, in research studies focusing on places, the researchers must try to generate images that let emerge the particular characteristics of the place, in order to stimulate a variety of narratives (Smith and Burch, 2012). Since our study involved two different - albeit connected - places, presenting sets of images that simultaneously depicted both of them helped us in making the citizens of both Gorizia and Nova Gorica feel part of the study. The interviewees could choose which image to comment on first, talk about the two photos and choose whether to compare them or not based upon their personal criteria. If we had showed a single photo related to Gorizia or Nova Gorica, we could have generated a situation in which we asked to the interviewees to express their thoughts, memories and observations regarding an image that could be little known or even not known. We could have embarrassed the interviewees denying them the opportunity to respond, making them feel uninformed and/or inadequate in relation to the expectations of the researcher.

46 So, Battello and I thought that the bipolar photo elicitation was the best tool to gather and understand the opinions, memories and different points of view of two different social groups. In fact, this tool has proved further valid when we discovered that the social groups to be interviewed were three: we became aware of the existence of the Slovene minority living in Italy during the research. A fortiori the bipolar photo elicitation helped us to communicate with this third social group that is part both cultures and territories considered, offering them an immediate feeling of having the opportunity to speak of their possible ‘double’ sense of belonging.

47 Second: increasing the equilibrium between the weakness and the structuring of the stimulus. One of the intents of the photo elicitation (as well as other qualitative techniques) is to reduce the level of directivity and intrusiveness by the researcher; with low directivity I refer to the possibility for the interviewee to decide the organization and content of his/her answers (Bichi, 2002) and with low intrusiveness I refer to the possibility to minimize the interventions by the researcher (Corposanto, 2004). As I already wrote above, often, the power of the image in an interview setting lies in the ‘weakness’ of the stimulus showed and in its polysemy. Similarly, Cronbach (1960) defines a situation such as non-structured when it has so few or vague meanings that you can assign to it almost any meaning and it is structured when it has a definite meaning for all subjects. The interpretation that a person assigns an image is therefore strongly influenced by his/her interests, values and, of course, knowledge; but, if a photo is ‘too’ difficult to interpret or it could be interpreted in ‘too many’ ways, it could also be misleading for the interviewees with respect to the focus of the research. Where this happens, the researcher is forced to intervene in the interviewees’ narrative instead of leaving them free to express themselves.

48 When I repeated the same test above with photo 1 and 2 with people who don’t have any problem with alcohol, they explained to me that if a researcher had shown them only the photo 2 on domestic violence, they would not have thought of alcohol. There is a specific reason: nowadays in Italy questions related to femicides and violence against women are much debated because they are serious emerging problems. Therefore, my respondents would not focused their comments connecting these to the alcohol if I had not explained that alcohol was the object of the research. Interviewees would have interpreted the problem depicted in photo 2 in a completely different way from the meaning that it had for the researcher. The latter would have had to intervene to drive the interview toward the central theme for research.

49 Photo elicitation, in general, concentrates on the subjective meanings that images have for the interviewees, but the image proposed does not always lead the researcher to the outcome he is expecting (Clark-Ibáñez, 2004) or can generate data that had been ignored until then (Schwarz, 1989). That is what happened to us in the research on Gorizia and Nova Gorica: some of our interviewees did not even recognise the plaque in photo 5 but we but we overcome this problem - without intervening on the interviewees’ narrative - thanks to the photo 6. From the interviews it emerged that when, in 2004, the plaque was installed in Transalpina square, for the inhabitants of Gorizia and Nova Gorica the border was already open; only a small number of interviewees affirmed that this event was meaningful, while others stated they have never seen the plaque. According to many of them, the plaque was installed as a ‘touristic attraction’ and in fact, for the out-of-towners, is one of the places to visit. Presenting photo 5 with 6 - which instead provoked the interviewees’ emotions and memories - allowed us to focus the narrative on the border in the past and in the present. If we had showed only the photo of the plaque, we would have had to intervene to explain what was our interest; instead, combining it with the photo of the old border, there was no need to intervene.

50 The increase of the equilibrium between the weakness and the structuring of the stimulus through the presentation of two photos, also allows a reduction of the possible error of the researcher resulting from his/her etic point of view and his/her subjectivity in the choice of the photos.

51 Third: a possible increase of consistency in the analysis phase. When the aim of the researcher is also to examine the attitudes, opinions or values ​​of interviewees in relation to a specific topic or object of research, it is possible use the bipolar photo elicitation to explore not only the position of the interviewees along a continuum, but also to check the consistency of their responses. If each set of photos is significantly combined with the other sets, in such a way as to obtain answers that can be cumulable (Guttman, 1944; Guttman, 1950), we can define as consistent the cases in which the interviewee shows the same attitude or opinion regarding all the sets. For example, in the research on Gorizia and Nova Gorica, if the interviewee shows a certain indifference to all photos of the city in which he/she does not live, it is evident that he/she little knows it and/or he/she does not feels a sense of belonging to it. Or, in research on alcohol, we can pay attention if the interviewee always expresses (or not) disapproval in front of the photos that correspond to the abuse or misuse of alcohol.

52 Therefore, besides analyzing the answers of the interviewees related to specific research questions (e.g.: could the education campaign have success? Despite the opening of the border, would there still be an emotional border?), it is possible to set up every couple of photos in such a way that they can be placed along a further continuum of consistency: when an interviewee expresses the same attitude and the same opinions through all the sets, his/her answers can be defined as consistent (Russo and Vasta, 1988).

53 To conclude: given the reflections that emerged from the use of the bipolar photo elicitation in the case studies that I present here, I believe that this technique is a very valuable tool when the research object can be explored, basically, from two perspectives, when the object of study can be examined through a comparison, and when the researcher wants to understand or investigate opinions, attitudes and/or perceptions organized around two poles that he/she is unable to perceive directly. Identical and different, positive and negative, past and present, legal and illegal, us and them: two faces of the same coin.