Abstract
This paper draws upon digital recordings of Northern Ireland football fans singing in the stadium during all 10 qualifying matches for the 2016 UEFA European Football Championship. Supplemented by participant observation and interview data with 21 supporters themselves, the paper challenges assertions within the literature which focus upon the predominance of sectarian singing amongst a section of Northern Ireland football supporters. Although vocal manifestations of football fandom may initially appear to be randomly driven by irrational emotions, on the contrary, there is an underlying structure and sequence to fandom in the stadium in which certain factors promote collective singing at particular times. The paper identifies four key themes in particular: the timing in a match; whether or not a goal has been scored; if there is a lull or a break in play; and the use of musical instruments to encourage the wider collective to sing. We argue that it is important to understand the process by which collective singing occurs in the football stadium rather than fixating upon the alleged racist or sectarian psychopathology of the individuals involved. Such knowledge may assist in supporting those fan organisations that seek to challenge discriminatory behaviour in the stadium, particularly in the current context of the European (UEFA) and World football governing bodies (FIFA) punishing fans collectively, regardless of whether or not the majority in the stadium are opposed to what is being sung in their name.
Aside from some excellent studies on the contribution of crowd behaviour to violence and football ‘hooliganism’ (Marsh et al., 1978; Stott et al., 2007), to date there has been little systematic sociological effort to document what a crowd of football fans sing during matches, when and what factors are ‘causally efficacious’ (Bhaskar, 1975) in encouraging them to do so. There has however been some work within anthropology and ethno-musicology on the role collective singing in the stadium can play in increasing tension and symbolic violence between rival supporters (Jack, 2013) and promoting actual physical violence, homophobia and racism (Herrera, 2018). Herrera’s work in particular highlights the process of deindividuation where being part of a crowd allows some fans to engage in utterances that they might otherwise refrain from using. This is an important observation; although as research recently published within this journal has suggested (Bell et al., 2020) being part of a larger crowd at a football match can also in fact lead to more moderate supporters feeling empowered to challenge racism and sectarianism. Despite this more contemporary work, our knowledge and understanding of the process by which songs develop within the football stadium remains underdeveloped. This is particularly significant given that the current policies of both the Union of European Football Associations (UEFA) and the International Federation of Association Football (FIFA) to close parts of stadiums (or the whole stadium) in the event of ‘offensive’ fan behaviour (typically a euphemism for racism, sectarianism, homophobia or xenophobia) are generally predicated upon vocal utterances emanating from within the crowd (see UEFA, 2016).
By improving our understanding of how songs and chants emerge in the stadium, football fans and governing bodies can be better equipped to challenge such behaviour from within. Such an approach focusing on in-stadium behaviour may assist the efforts of organisations which have been established to specifically address various forms of intolerance in the game. In the UK, the most notable of these organisations are ‘Kick it Out’ and ‘Show Racism the Red Card’ (Dixon et al., 2016). In a wider context, ‘Football Against Racism Europe’ (FARE) focuses upon networking, organising media campaigns against various forms of intolerance and general awareness raising of discriminatory behaviour in the game.
But while undoubtedly a laudable enterprise, the effectiveness of such educational and awareness raising work on its own has been questioned, both by academics (Dixon et al., 2016; Garland and Rowe, 2014) and fans themselves (Cleland and Cashmore, 2014). There are two points to be made here. Firstly, as Dixon et al. suggest, it is simply too much to expect one-off workshops or promotional displays to change the attitudes of a critical mass of individuals given that racism, sectarianism and other forms of prejudice are societal wide issues, a point acknowledged by Bairner and Darby (1999) in the Northern Irish context. Secondly, educational approaches do not directly address fan behaviour in situ and tend to ignore the significance of the spatial distribution of supporters in the stadium, which is crucial to creating the ‘atmosphere’ at a football match (see Dixon, 2012).
By drawing upon the case study of fans of the Northern Ireland football team, where there were historic issues associated with sectarian vocalisations emanating from a section of supporters (Bell et al., 2020), this paper offers a sociological explanation of fandom that may have policy implications given current draconian measures to close entire stadiums which punish football fans en masse for the actions of a small number of individuals within their midst.
‘Shall we sing a song for you?’
Other than the efforts of Marra and Trotta (2019), who recorded the crowd at 21 Clube Atlético Mineiro matches between 2011 and 2015, there have been few attempts at recording and analysing the fan collective in the football stadium, although the method has been used in a study of Swedish ice hockey fans (Granström, 2011). Drawing upon participant observation and recording of the crowd at two high-risk Swedish ice hockey matches, Granström developed a ‘taxonomy of cheers’ and found that the content of ‘cheers’ can be taken as a sign of the social identity shared by supporters. At times a small number of individuals attempted to start aggressive ‘cheers’ and provoke a negative social identity among the crowd, but crucially ‘cheerleaders’ within supporter groups were able to stop this by introducing supportive ‘cheers’ which promoted a peaceful social identity.
This is in line with much of the work on ‘Ultra’ fans in continental Europe (or barras bravas in South America) who use song-leaders (capos) and co-ordination in the stadium itself to sing (Giulianotti, 1999). Standing with their back to the game, ‘some Ultra leaders employ microphones or small amplifiers to lead the chants’ (Giulianotti, 1999: 54), to help create an ‘atmosphere’ in the stadium (Dal Lago and De Biasi, 1994).
But it is important to understand the heterogeneity of ‘Ultra’ groups across Europe (Doidge, 2015). Different sub-sections amongst supporters, even those from the same club, may promote a differing ethos; some of tolerance and respect, others of suspicion and hostility towards minority groups. Hodges (2016) conducted participant observation of NK Zagreb supporters in Croatia, and found that while members of the ‘Bad Blue Boys’ (BBB) tended to engage in sexism, racism and homophobia in the stadium, members of a differing ‘Ultra’ group, the ‘White Angels’ (WAZ), promoted tolerance and respect for others. Similarly, at Olympique Marseilles, Assollant et al. (2011) found that the ‘Commando Ultras’ adopted a conservative, pro-French, anti-immigrant and ethnocentric ethos while the ‘South Winners’ developed a multi-cultural identity and affiliation for Marseille over and above France.
While ‘Ultra’ supporters across Europe and beyond have been congregating together in particular parts of the ground to create an ‘atmosphere’ for many years, in a UK context, informal supporter groups styling themselves on ‘Ultra’ practices have emerged more recently. This includes groups such as the ‘Holmesdale Fanatics’ at Crystal Palace and the ‘Green Brigade’ at Celtic FC. Beyond their efforts to promote collective singing and holding political displays, former Celtic player and manager Davie Hay has criticised the latter group for their more illicit behaviour during matches, including the unveiling of a banner with ‘Hun Scum’ written on it in September 2016. 1
However, the ‘Green Brigade’ themselves would refute any allegations of sectarianism, referring to themselves as ‘anti-fascist’ and ‘anti-sectarian’ (Lavalette and Mooney, 2013), with the singing of pro-IRA songs viewed as a legitimate expression of Irish republicanism which the Scottish government tried to ‘criminalise’ through the introduction in 2012 of legislation banning ‘offensive’ singing (Millar, 2016). This argument is relatively well supported by academics and appears to follow that which Armstrong (1998) made in relation to the Football Offences Act (1991): that football fans are being ‘criminalised’ in an ‘age of intolerance’ (Waiton, 2014). The implication here is quite simple; those who are intolerant are not those singing about ‘taigs’ or ‘huns’ 2 (or for that matter ‘Yids’ when Spurs are playing, see Poulton and Durell, 2016) but those who have such low levels of tolerance that they are offended by such behaviour.
Yet it is our contention that not only does such a position shift the blame onto the victim of verbal abuse and ignores the philosophical argument in terms of ‘other regarding’ actions (Mill, 1859); but it engages in a rather patronising cliché that such behaviour is just what the ‘working-class’ ‘do’ at football. Thus, Lavalette and Mooney (2013: 24) assert that ‘the SNP Government are advancing a model of a “new” “modern” Scotland in which aspects of working-class culture are seen as highly problematic’ (see also May, 2015). The challenge with such an argument (beyond its pejorative portrayal of the ‘working-class’ generally) is that it focuses so much on who (in terms of individual/communal pathologies) is responsible for such discriminatory behaviour that it neglects to analyse how such behaviour occurs. Providing such an analysis is one of the key purposes of this paper.
Football in Northern Ireland
By 1969 civil disturbances in Northern Ireland would turn to violence as the ethno-national conflict between Irish republicans, Ulster loyalists and the British state, euphemistically referred to as the ‘Troubles’ began, only ending officially in 1994 with the paramilitary ceasefires, after the deaths of more than 3600 people (McKittrick et al., 1999).
It has been argued that a change occurred in the atmosphere at Northern Ireland matches in the wake of the Anglo-Irish Agreement in 1985 and continuing political violence of the ‘Troubles’ (Sugden and Bairner, 1993). From this point onwards, football became a significant symbolic battleground for Protestant and British ‘Ulster’ identities in the context of a political conflict with Irish republicanism (Bairner, 1997). Thus, a section of Northern Ireland supporters increasingly began to sing songs with anti-Catholic sentiments, such as the ‘Billy Boys’, which contains the line ‘We’re up to our knees in fenian blood’ (Bairner, 2005). This resulted in the ‘colonization of the national team by a vocal minority of openly sectarian supporters who often diverted attention from football and focused it instead on anti-Catholic vitriol’ (Hassan et al., 2009: 749). Using Jennifer Todd’s (1987) typology, Bairner (2005) argues that the identity most clearly expressed at Northern Ireland football matches is an ‘Ulster Loyalist’ as opposed to a more ‘moderate’ ‘Ulster British’ one. The highest profile incident in terms of sectarian fan behaviour came in 2001 with the treatment of Glasgow Celtic player Neil Lennon who was booed by a section of the home crowd every time he touched the ball in a match against Norway (Reid, 2008). 3 By August 2002 after more sectarian behaviour from a section of his own fans, Lennon withdrew from the Northern Ireland squad and retired from international football due to a death threat he received, allegedly from the Loyalist Volunteer Force (Reid, 2008).
Although efforts to eradicate sectarianism from Northern Ireland matches predated the ‘Neil Lennon incident’, this development led to an intensification of partnership efforts between the Irish Football Association and the Amalgamation of Official Northern Ireland Supporters’ Clubs (AONISC). Fans continued with their ‘Football for All’ campaign to challenge sectarianism at the match and created a new canon of non-political songs, such as the rather tongue-in-cheek ‘We’re not Brazil, We’re Northern Ireland’ and ‘Green and White Army’ to replace sectarian songs such as the ‘Billy Boys’, or more specifically Ulster loyalist songs such as ‘The Sash’. Fans were also encouraged to wear the team colours of green and white rather than the red, white and blue associated with Linfield, Rangers FC and the Union Flag of the UK to which some fans had been symbolising their allegiance. In 2006, Northern Ireland fans, or the ‘Green and White Army’ (GAWA) as they had otherwise become known by then, were awarded the Brussels International Supporters Award by the governing body for football in Europe (UEFA) in recognition of their efforts to challenge sectarianism at Northern Ireland matches (UEFA, 2009).
The transformation of the Northern Irish international football fan culture away from sectarianism in recent years and the specific details of the ‘Football for All’ campaign has been documented recently within the pages of this journal (Bell et al., 2020). Yet despite this, the most recent work on Northern Irish football fandom (Keown, 2020) suggests that Protestant Northern Ireland supporters attend matches, not only to support their team, but also to express the ‘superiority’ of their identity over their northern Catholic counterparts. Yet as (Bell et al., 2020) argued was the case with the literature on Northern Ireland fans more generally, there is little empirical evidence to support this analysis.
This paper will draw upon empirical data from 10 Northern Ireland matches to assess whether sectarian songs are being sung in the stadium. If not, what has replaced them and how do such songs emerge within the fan collective? This latter point is an important one in a wider context as it may provide useful information on fan behaviour for other countries where forms of intolerance remain pressing issues within the confines of the stadium itself, including Scotland where the Offensive Behaviour At Football and Threatening Communications (Scotland) Act 2012 was repealed on 20 April 2018.
The remainder of the paper is structured as follows. The following section highlights our chosen methodology underpinning the research. The paper then proceeds to discuss the empirical data which reveals that there is an element of underlying structure and sequence to fandom in the stadium. Finally, we offer some concluding remarks with regards to the significance of the research in policy terms, particularly with regards to the current disciplinary regulations imposed by UEFA and FIFA on national associations and clubs for ‘offensive’ behaviour perpetrated by their fans at matches.
Methods
Data collection
This paper is based upon data from three sources: participant observation of the behaviour of Northern Ireland fans in the stadium during all 10 European Championship qualifiers (otherwise referred to as EURO 2016) held between September 2014 and October 2015; digital recordings of the songs of the crowd across the 90 minutes of each match; and interviews with Northern Ireland fans themselves to explore the emerging findings in greater detail (n = 21). The data referred to in this paper were part of a wider research programme which aimed to investigate contemporary patterns of Northern Irish international football fandom (Bell, 2017). This research itself was a result of the lead author’s prior engagement at the Institute for Conflict Research in Belfast, critically evaluating the efforts of sports organisations involved in good relations work.
Matches were recorded on a digital recording device which was switched on c. 15 minutes prior to kick-off in each match and similarly was turned off c. 15 minutes after the match ended. The sub-section on data analysis will shortly outline how the match recordings were transcribed and how songs were counted, particularly taking into account the differential positioning in the stadium and the impact of this on standardising the counts.
The extent of researcher participation in fan practices would have involved standing for national anthems at matches and applauding a good piece of play or a goal. But the researcher did not sing with them, drink (alcohol) with them and try to befriend them as is often the case in research upon football fans (see Spaaij and Geilenkirchen, 2011). The extent of researcher participation is located within Bryman’s (2012) six-fold participating-observer role typology, somewhere between the ‘minimally participating observer’ and the ‘non-participating observer with interaction’.
With regards to the interview data, the selection strategy underpinning the recruitment of interviewees was purposive and involved ‘maximum variation’ sampling (Denzin and Lincoln, 2005). Most supporters were male (19) and Protestant (18), although the cohort also included female fans (two), Catholic fans (three), and a wide range of ages (from 21 to 67 years old). Ten Northern Ireland supporters’ clubs were contacted and asked if a member would be interested in participating in the research. Semi-structured interviews were conducted between January and February 2016 in the aftermath of the EURO 2016 qualifying campaign.
With prior informed and written consent, all interviews were digitally recorded. Interviewees were assigned a pseudonym to protect their identity, and these are used throughout this paper.
Data analysis
Digital recordings of the crowd at each match were transcribed in detail using SpeechExec software (complete with foot pedal). As was the case with the interviews, the match recordings were transcribed verbatim, along with timings of when shouts were made by fans (as individuals, sub-groups and as a larger collective) and when songs started and finished. While everything audible on the recording was transcribed (individual fan, sub-group and wider fan collective), only those songs which were sung by a critical mass of the crowd were counted as collective crowd efforts and included in the final analysis. This was to standardise the song counts from different locations in the stadium and ensure greater reliability of the data. For example, at Windsor Park in Belfast several thousand fans typically sing in unison (of the approximate 10,000 in attendance). These are counted as crowd-wide renditions. Similarly, away from home, if for example 200–300 of the 500 away fans in Greece were singing this was counted (instances where small groups of 10 or 15 fans were singing on their own were not). In other words, Table 1 documents songs which were sung by a large section of the Northern Ireland match crowd both home and away.
Most popular five songs sung during EURO 2016 qualifying matches.
NVivo 10 was utilised to code, organise and analyse the interviews and match transcripts. However, despite attempts to improve data validity, all interviews lead to data which is co-constructed through dialogue between interviewee and interviewer (Rubin and Rubin, 2005). It must also be recognised that case-study data is context dependent and this impacts upon external validity (Yin, 2009). Yet while the specific context of Northern Irish international football fandom must be kept in mind, there are enough similarities in football fandom worldwide to suggest that the findings of this paper are not redundant for other national contexts.
Research findings
‘Ulster Boys’ or ‘Billy Boys’?
Table 1 lists the five most popular songs sung during the EURO 2016 qualifying campaign by the wider Northern Ireland fan collective in the stadium.
The match data indicates that across all 10 matches, Northern Ireland fans in the stadium sang as a collective a total of 604 times, an average of just over 60 songs per game. As the table indicates, the two-line ‘Green and White Army’ chant was the dominant song sung by Northern Ireland supporters. On its own it accounted for more than a third of all songs sung during the campaign (34.6%) and averaged 20 renditions per match. In nine of the 10 matches it ranked as the most dominant single song by some considerable distance. As Table 1 also illustrates, the five most popular songs alone accounted for 63.9% of all the collective vocal efforts of the Northern Ireland fans in the stadium across all 10 matches. Although supporters drew from a repertoire of 50 songs across the entire qualifying campaign, most of these songs were only sung on several occasions. The patterns of GAWA fandom which were established through the ‘Football for All’ campaign from the early 2000s onwards are recursively (re)constituted across time and space in routinised social praxis (Giddens, 1984), wherein a small number of specific songs are sung repetitively.
What is clear from the match data is that vocal manifestations of ‘Britishness’ in the stadium are not (at least in the contemporary period) as dominant as the literature has suggested, only providing the focus for 3.1% of collective crowd songs. Perhaps even more significantly, the assertion that a section of Northern Ireland fans sing sectarian and anti-Catholic songs in the stadium is not corroborated by empirical evidence in the contemporary period. One issue which was striking from both the participant observation and match recordings was that even sub-group songs within the stadium tended to steer clear of sectarianism, although this does not mean that sectarianism has been eradicated from Northern Irish international football fan culture (and there can still be some small groups of supporters who engage in such behaviour post-match in the pub, see Bell, 2017). Rather, it means that there are certain ‘moral norms’ (Giddens, 1993) which have been established as to what behaviour is expected of the GAWA within the stadium itself.
However, there are two points to consider here. Firstly, the five other teams Northern Ireland played during the qualifying campaign for EURO 2016 were the Faroe Islands, Finland, Greece, Hungary and Romania. There is no history of constitutional conflict or political antagonism between Northern Ireland and any of these nations. Yet should Northern Ireland play the Republic of Ireland in a competitive match, the political context would be much different and within the parameters of ‘Social Identity Theory’ (Tajfel and Turner, 1979), this could impact on the interactions and the singing between the two sets of supporters. Secondly, as the wider literature on ‘Ultra’ fans suggests, there is a heterogeneity within the support base of Northern Ireland. While the ‘Green and White Army’ social identity remains the dominant one which is promoted by most fans on match days, there remains a section within the support base for whom ‘Ulster’ specifically means ‘Protestant Ulster’ (Shirlow and McGovern, 1997). For these fans, supporting the Northern Ireland team provides a public platform from which to visibly affirm their commitment to the Union (Sugden and Bairner, 1993). This is particularly the case with regards to the behaviour of a section of supporters during the playing of ‘God Save the Queen’ at matches and the occasional singing of tunes such as ‘Ten German Bombers’ (with the line ‘And the RAF from Ulster shot them down’). There is also a complexity to some of the songs which Northern Ireland fans sing. This is particularly the case with ‘Everywhere we go/Ulster Boys’. While for most fans, the singing of such a song illustrated their desire to simply have a ‘good time’ in supporting the team and they used ‘Ulster’ as a synonym for Northern Ireland (Protestant and Catholic alike), other fans employed the term ‘Ulster’ more narrowly as representative of an ‘Ulster Loyalist’ identity (Todd, 1987).
The fifth most popular song sung by Northern Ireland supporters across the EURO 2016 qualifying campaign was ‘the bouncy’. To those not familiar with the practice, the sight of grown men and women, often with their arms around one another, repeatedly jumping up and down in unison while singing, ‘Let’s all do the bouncy, let’s all do the bouncy,’ is perhaps one of the stranger sights in football. For members of the GAWA, ‘the bouncy’ is the ‘best craic’ had at Northern Ireland matches and is an expression of ‘pure joy’ to indicate their support for the team: Just a few people who have had a few too many drinks like, just starting! [laughs] Or it’s usually not long after you score and you are really pumped up on adrenaline and all of this here and its sort of a good way to burn off a bit of energy kind of thing like. And you just jump up and down. So someone will usually start singing that and jumping up and down and if there is something to be really happy about or like a continuation from a celebration it seems to be, so if they score or save a penalty or something – I can’t remember the last time that happened, but that’s usually when you get a sort of continuous celebration. And it looks really good, for the players on the pitch I’m sure, and for the tv cameras and stuff. (Josh)
Yet several alternate versions of ‘the bouncy’ are performed by some sections within the fanbase of Rangers FC. One version of the Rangers ‘bouncy’ includes the line, ‘If you cannae [sic] do the bouncy, you’re a taig’. This is a directly sectarian version of ‘the bouncy’ which elements within the tabloid media in the UK have suggested represents fans’ desire to jump upon the head of a Catholic, and more specifically, Robert Hamill who was brutally murdered by a loyalist mob in Portadown in 1997.
4
While many Northern Ireland supporters were completely oblivious to this alternative interpretation of ‘the bouncy’, some fans had come across such views before: Like I know you have like the ‘bouncy’, or things like that. And [pause], I never really knew this to lately [sic], but I was reading, just reading a few things on the internet and like there’s quite a lot of things, a lot of people take that as being very sectarian. . . apparently a lot of people relate that back to, I think it was a republican guy who was murdered, like people stamping on his head. But I never, I was just reading that lately, but I would never have thought of that in a million years. It’s just a party song. It’s a bit of craic. Similar really to any, you think of some of those countries even where they turn their back and they just bounce. (David)
The assertion by some tabloid journalists in the UK that those fans who do ‘the bouncy’ do so to mimic a murderous lynching is not only factually incorrect (as the practice long predates the 1997 murder of Robert Hamill) but it ignores that fans at some clubs in England and Scotland engage in similar practices.
5
It is also to commit a lay version of what Bourdieu (1988) refers to as the ‘scholastic fallacy’. This is where the actions of a group are interpreted by an external observer with reference to their own habitus and value system without any relation to the meaning of a practice for the actual group under study themselves (Bourdieu and Wacquant, 1992). It is also to ignore Durkheim’s (1912: 283–284) point about the behaviour of those involved in collective religious rites, that: we may miscalculate when we try to assign each gesture a precise purpose and a well-defined rationale. There are some that serve no purpose at all; they answer simply to the worshippers’ feeling that they need to act, to move to gesticulate. They may leap, turn, dance, shout, and sing, and this agitation may have no discernible meaning.
While some of the internet suggestions as to the origins of the ‘bouncy’ (if applied to Northern Ireland fans) were ones that many supporters found offensive when it was pointed out to them, Northern Ireland fans should perhaps reflect upon the potential harm which differing interpretations of the same gesture could cause (even unintentionally) to the efforts to challenge sectarianism and promoting ‘Football for All’.
‘And this is why we sing. . .’
Within the wider literature on football fandom, several studies have suggested that collective singing is more popular away from home, as this is where the most ‘dedicated’ supporters who are willing to travel far and wide to support their team are to be found (Brown, 1998; Pearson, 2012). Yet the data from this study suggests that there is not a simplistic home/away dichotomy wherein away fans sing more so than those at home. There are other more significant factors which influence collective vocalisations at matches besides the home/away distinction. Although these ‘generative mechanisms’ (Bhaskar, 1975) on their own can improve the likelihood of a crowd song emerging, they are not mutually exclusive and may overlap with one another to provide the optimal conditions in which a collective vocal effort may be sustained. They include:
the period at the beginning/end of a match;
if Northern Ireland score/concede a goal;
if Northern Ireland are playing well and there is a break in play (e.g. corner or free-kick); and
if a musical instrument (e.g. drum) supports the attempted song.
The opening and closing 10 minutes of the game are key periods in which a sizeable proportion of collective GAWA singing is concentrated: Usually you don’t pay that much attention to the football for the first 10 minutes unless something happens! [laughs] You are just going through the repertoire of songs as such and then like I say you get a bit of action or something and like a specific player scores or does something good, there’s a song for them, or if the referee in our opinion isn’t having a particularly good game, you give him stick, this kind of thing. (Josh)
This tends to be reflected in the match data. In the first nine minutes of the home match against Hungary the GAWA sang collectively 12 times (36.4% of the songs sung in the first half). At home to Finland the GAWA sang eight times in the first four minutes alone (taking another 10 minutes to sing a further eight songs). Nor are such examples restricted to home games. In the away match against the Faroe Islands, the GAWA sang as a collective 11 times in the opening 10 minutes of the match (36.7% of all first half songs).
A similar tendency towards collective singing emerged towards the final whistle in matches. Although the Northern Ireland team lost 2–0 in Romania, the GAWA sang collectively four times in the four minutes of injury time at the end of the game. To put this in context, they only sang collectively three times in the previous 17 minutes of the match. In Helsinki, the final match of the campaign when the team had already qualified for the final tournament in France, in the last 10 minutes the GAWA sang 11 times (40.7% of all second half songs).
The two matches in which the GAWA sang most frequently as a collective were the away win against Greece (83 times) and the home win against Finland (86 times). The first half of football against Finland at home wherein Northern Ireland scored three goals (albeit one was disallowed) also contained the largest number of collective songs sung in a single half in the entire campaign (49 songs). Indeed, Northern Ireland fans sang as a collective after all 16 goals the team scored in the campaign, typically within a minute of a goal being scored, and often as quickly as 20–30 seconds after a goal. Conversely, the match in which the GAWA sang the least was the away match in Bucharest, wherein Northern Ireland did not score a goal and lost their only game of the entire campaign.
Yet not only did Northern Ireland fans collectively respond when their team scored a goal, as Jamie suggested, ‘they would sing, sort of straight away, if they conceded for a while to try and lift them again’. Of the eight goals that Northern Ireland conceded in the EURO 2016 campaign, the GAWA sang as a collective immediately after six of them. The two matches in which they failed to vocally respond were when Romania went 2–0 up in Bucharest and the game was all but over and when Greece pulled a goal back in the 86th minute in Belfast to make it 3–1, at which point Northern Ireland were practically assured of qualifying for the final tournament in France regardless.
‘Where it began. . .’
Football fans at the match are in attendance both as individuals and as part of a larger crowd. It is this complex interaction between agency and structure which must be grasped. As individuals, supporters influence the wider crowd (and fans would also argue that they can influence events on the pitch, see Marra and Trotta, 2019); but they are also liable to be influenced by the wider crowd through the ‘duality of structure’ (Giddens, 1984). Collective GAWA crowd songs at matches emerged in two distinct ways which are suggestive of this interplay between agency and structure. Firstly, individual fans or small groups of supporters in close bodily proximity to one another may attempt to start a song. Secondly, fans may join in with a song which has been started by the use of a musical instrument, in particular bass drums, but also including hooters, saxophones and make-shift ‘instruments’ (such as fans banging the stands with their hands and feet as occurred in Athens). For both individual/sub-group efforts and those prompted by a musical instrument, timing is crucial. For a song to take off successfully around the wider crowd, it needs to be the ‘right song’ at the ‘right time’: You could start the wrong song and it will not carry you know. . . the song needs to be right for the mood of where we are in the game. . . Because you’ve got 90 minutes of emotion, you are going ‘Argh are we gonna [sic] hang on?’ You know? Or ‘Yay!’ Or ‘Oh my God we’re useless!’ So ‘We’ll support you ever more’ is all you can do. And the rhythm of the songs, I don’t know whether they match your heartbeat or something, you know, it needs to be the right song for where you are emotionally watching the game. (Irene)
Attempted songs were also more liable to successfully take off if they occurred during a quiet moment in the match, or in particular, a break in play when supporters were not so intently focused on the action on the pitch, such as a free-kick or corner-kick. The match data supports this view with regards to singing the ‘right’ songs at the ‘right’ times. Consistently across all 10 matches, attempts at songs during loud periods of the match or when the crowd were intently watching the action on the pitch failed. Conversely, during a lull in play, even a lone voice from the crowd can, within a few seconds, start several thousand fans singing in unison: A few people start it and then very quickly other ones catch on and a lot of us are sheep and just follow, you know! So, if we hear people starting a song up and we know sort of the words of it, we have heard it a hundred times, people just generally join in. (Kyle)
Fans used terms such as ‘rippling out’ (Gareth and Paul) and ‘infectious’ (Tony) to describe how songs such as these take off around the crowd in the stadium. However, individual fans attempting to start a song on their own are more likely to fail than they are to succeed. Small groups of supporters are likely to have more success; but it is the use of a musical instrument which is most influential in determining not only what the crowd sing, but whether or not the song is likely to take off in the first place and be sustained: Cos you have the beat and people will start that whether it’s ‘Green and White Army’ or whatever, so the drum is the one that would start it and normally control what’s sung you know? (George)
But how can the beat of a drum dictate or control what song fans will sing? Anthony Giddens (1984) highlights the significance of ‘practical consciousness’ in structuration theory. This is the tacit knowledge that comes from bodily experience in the world, from routine and repetition, the ‘taken for granted’, wherein individuals ‘just know’ how to behave in a particular ‘form of life’ (Bourdieu, 1977; Wittgenstein, 1953). Through attending matches, Northern Ireland fans learn the practices and songs associated with GAWA fandom (see also Dixon, 2012). Suggesting that with regards to the atmosphere at a Northern Ireland match, ‘You just feel it in your bones,’ Josh continued: Finaghy and Armagh supporters’ clubs bring bass drums to the matches, which sound really good ’cos then they can beat that to the tune of a song and that keeps everyone in rhythm or they can start a song using the drum, like, if the drum will start the speed then people will know the song and then start songs that way. (Josh)
Josh’s point can be taken both metaphorically and literally as the noise reverberates around the ground, and at times, up through the floor of the stadium and into the body. Particular beats on the drum are effective in ‘encouraging’ the majority of the crowd to sing, with the instrument acting as a mnemonic device for fans to act in unison. It is the ‘mutual focus of attention’ in the stadium which builds ‘emotional energy’ amongst fans (Collins, 2004: xii), and which ‘permits the group’s harmonious movements, these gestures and cries are inclined to be rhythmic and regulated, and become chants and dances’ (Durkheim, 1912: 163). The ‘communicative function’ of the gesture is particularly relevant (Mead, 1934); two slow drum beats means fans are to sing ‘Green and White Army’, five beats leads in to ‘Northern Ireland’, nine quick beats requires fans to shout out ‘Ulster’, while quick single beats in succession followed by a pause begins ‘Everywhere we go/Ulster Boys’. The meanings of these gestures ‘are not subjective, not private, not mental, but are objectively there in the social situation’ (Mead, 1934: xxi). It is social praxis and the routine and repetition associated with fan culture in the stadium which promotes knowledge of the songs and appropriate responses to these gestures (Giddens, 1984).
Conclusion
Primarily drawing upon recorded match and participant observation data from the qualifiers for the 2016 UEFA European Championship, this paper has argued that it is more accurate in the contemporary period to refer to the ‘Ulster Boys’ rather than the ‘Billy Boys’ dominating the atmosphere at Northern Ireland matches. While Northern Ireland fans as a collective no longer sing sectarian songs in the stadium, the vocal point of reference also tends to have shifted towards themselves (GAWA), Northern Ireland or ‘Ulster’ rather than ‘Britannia’.
The heterogeneity noted in the wider literature on ‘Ultras’ is also apparent within the Northern Ireland fanbase. It is a truism that some fans do still adhere to a narrow ‘ethnic’ identity which symbolically draws upon the iconography of Ulster Protestantism and the connection with the rest of the UK in the contemporary period; 6 but others support the team because they are proud of being ‘Northern Irish’ and do not endorse displays of ‘Britishness’ at matches. Such complexity within the fanbase has been overlooked in the literature, which has also marginalised the ludic aspects of fandom which are central to GAWA fan culture and which focus upon humour and self-deprecation to promote solidarity amongst the fan collective and challenge historic perceptions of supporters as ‘Super Prods’.
The paper has further argued that although from the outside football fandom may appear to be driven by irrational emotions and appear as a random set of outbursts of ‘collective effervescence’ (Durkheim, 1912), on the contrary, there is an underlying structure and sequence to fandom in the stadium, an ‘unwritten musical score’ (Bourdieu, 1977), in which certain factors are ‘causally efficacious’ (Bhaskar, 1975) in promoting collective singing at particular times. Four key themes in particular were identified: the timing in a match; whether or not a goal has been scored; if there is a lull or a break in play; and the use of musical instruments to ‘encourage’ the wider collective to sing. The fact that five songs alone accounted for 63.9% of all collective GAWA vocal efforts across the 10 matches indicates that continuity, through repetition and routine, is also very important in sustaining a ‘social system’ across time and space (Giddens, 1984).The bass drum and other musical instruments act as an effective ‘mnemonic device’ which promotes memory recall and the continued singing of songs such as ‘Green and White Army’ more than 15 years after they were introduced into the fan repertoire.
If there are external factors which influence the fan collective on what to sing and when, it stands to reason that there are lessons which could be learnt for football contexts where a section of supporters are dictating the atmosphere in the game and propagating racist or xenophobic singing. Supporting fan groups opposed to such behaviour should be the aim for UEFA and FIFA, as well as national associations, to challenge it more effectively in the longer term. The current UEFA and FIFA policy of stadium bans for an entire fanbase based upon the actions of a small minority of supporters not only appears to be unfair, but the effectiveness of such policies remains unproven.
On 1 June 2013, UEFA (2016) implemented new regulations which provided for the full or partial closing of football stadiums in the event of ‘offensive’ fan behaviour. The regulations state that these policies can be implemented should ‘one or more’ fans of a club/team engage in ‘offensive’ behaviour within the stadium (UEFA, 2016: 13). Football’s world governing body, FIFA (2011) impose a similar penalty for such behaviour, with a comparably low threshold of involvement from supporters required to punish all fans collectively for the behaviour of a very small minority. As argued previously within the pages of this journal (Bell et al., 2020), clubs and national associations should engage in dialogue with their supporters and assess the potential to create fan-led initiatives which can challenge those engaged in discriminatory singing or chanting and promote more positive forms of football fandom. In the Northern Ireland case, musical instruments such as drums have become instrumental in the battle against sectarian singing and there is the potential to support such grassroots led attempts which would have a greater chance of long-term success in transforming a fan culture than any attempts which appear to be imposed from the ‘top-down’.
In attempting to document the contemporary vocalisations of Northern Ireland fans in the stadium, this paper has not sought to ‘airbrush’ history or the damage which the sectarian behaviour of a section of supporters historically may have had on patterns of football support and community relations in Northern Ireland more generally. But what we would argue is that it is more important to understand the process by which collective singing in the stadium occurs than it is fixating on who is responsible for it. To suggest that ‘offensive’ singing is in any way innate to working-class culture, as some of the critiques of the introduction of the Offensive Behaviour at Football and Threatening Communications Act in Scotland in 2012 appeared to suggest (Waiton, 2014), are essentialist and as damaging to those working-class communities they purport to defend than any government legislation. The case study of Northern Ireland fans has highlighted that singing at matches, whether sectarian or otherwise, is contextual and becomes routinised through praxis.
Yet while removing the worst excesses of sectarianism from Northern Ireland matches has been a major achievement on the part of Northern Ireland fans and the Irish Football Association, it has not depoliticised football fandom, and it still remains the case that the Northern Ireland team is one which predominantly receives support from the Protestant community while northern Catholics tend to look towards the Republic of Ireland (Hassan, 2005). This suggests that despite the almost total eradication of blatant sectarianism within the stadium itself, in the absence of any politically agreed alternative post-conflict state (as occurred in post-apartheid South Africa; see Bornman, 2006), there may be genuine limits as to the symbolic appeal of the Northern Ireland team across the community divide. While the ‘utopia’ of a football team supported by everyone in Northern Ireland may never quite be achieved, as Thomas More (1965 [1516]: 39) once wrote, at the very least ‘what you can’t put right, you must try to make as little wrong as possible’.
Footnotes
Acknowledgements
The authors would like to thank the two anonymous reviewers for their insightful comments on an earlier draft of the paper.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/ or publication of this article: The research drawn upon in this paper is based upon the lead author’s doctoral work. Living expenses were funded by a PhD studentship from the Department for Employment and Learning in Northern Ireland (September 2013-September 2016). Flights and accomodation for the five away matches in the EURO 2016 campaign were paid for by Ulster University.
