Community Impact of Public Processions

The study examined the community impact of public processions, with a particular focus on processions which were perceived to be 'problematic'. The research involved a multi-method approach and included qualitative and quantitative data collection. Although the research considered a wide range of processions (including community and political), its particular focus was on Loyalist and Irish Republican processions.


1. This study was commissioned as the community impact of marches and parades. We refer to public processions throughout, in line with current legislation (Police, Public Order and Criminal Justice (Scotland) Act 2006), and indicative of the broad range of events recorded under this category by local authorities.

2. Complete data was not available from all local authorities. Where data was missing, totals were extrapolated on the basis of available data and trends, see discussion in chapter two for more details.

3. These figures are based on local authorities where a full breakdown of procession types was available. Extrapolated local authority data was excluded. In 2012 for instance there were 2,280 such notifications (this comprised of 1417 community, 773 Loyalist Order, 45 political, 41 Irish Republican and 4 diversity processions).

4. Though a proportion of respondents in the pre-procession surveys were aware of some future procession in their area (ranging from less than 50% of respondents in Govan to nearly 100% of respondents in Coatbridge) the extent to which respondents were aware of the specific processions that were of interest to this study was impossible to quantify due to respondents own uncertainties as to dates and procession types.

5. We will refer to processions for the purpose of this report, in line with current legislation (Police, Public Order and Criminal Justice (Scotland) Act 2006), and to indicate the broad range of events that are recorded by local authorities under this category.

6. The Orr report distinguished between 'Orange' and 'Catholic' processions. However, as he acknowledged, the Orange Order is one of several loyal orders (which include the Apprentice Boys of Derry and Grand Black Chapter (also referred to as Royal Black Institution).

7. Figures produced by Strathclyde police show that in 2011-12,740 processions were organised by the loyal orders, compared to 33 Irish Republican processions.

8. See Tackling Sectarianism Advisory Group (2013).

9. The Scottish Social Attitudes report on public attitudes to sectarianism and the qualitative study on community experiences of sectarianism (both conducted in 2014) provide more recent evidence.

10. This includes 5% who did not attend or were not invited to a social event, 5% who believe they were refused a job or promotion and 7% who say they have been harassed or threatened because of their religious beliefs or background

11. Although the Orange Order proactively engaged with discussions concerning the 2014 Referendum in Scotland, and urged its members to take a pro-union stance.

12. This compares to football matches which accounted for 31% (267) of all charges in 2011-12 and 16% (109) in 2012-13. This drop is related to the introduction of the Offensive Behaviour at Football and Threatening Communications (Scotland) Act 2012 that may have dealt with offences which would previously have come under section 74 of the Criminal Justice (Scotland) Act 2003.

13. These figures only refer to recorded data.

14. In 2013, the United Nations Special Rapporteur (2013) considered Scotland (alongside other parts of the UK) in relation to rights to freedom of peaceful assembly and association, and criticised the use of a 28 day-period of notification in Scotland, urging that the legal framework be amended to 48 hour notice.

15. While organisations, both Loyalist and Irish Republican, may employ bands to participate in their processions, bands may be unattached to a specific organisation. On occasion, and increasingly so, individual bands will organise a procession which consists of a number of bands who may, but may not be, affiliated to a particular organisation. In this report for example, we refer to the Pride of Govan Flute Band procession, as well as the Irish Republican Bands Alliance.

16. Perth and Kinross provided incomplete data and so estimated total figures are based on these (rather than extrapolation from Orr report). Clackmannanshire provided lists of processions which allowed for the calculation of totals but insufficient detail re procession type. East Lothian also provided incomplete data, though the coverage of the data did not allow for estimated totals to be reliably produced, therefore figures for this area were based on extrapolations. Where recent figures were not available, 2003 figures for procession numbers in a given local authority were used to produce an estimate on the basis of the overall observed trend in procession numbers across Scotland up to 2013.

17.A key pragmatic constraint for the study was the timing of 'interesting' processions. The research timetable did not allow for Stage Three to commence until towards the end of June. In practice of course, many interesting processions occur in May and June.

18.On-street and business surveys only.

19. Each of the main Loyalist processing organisations has similar large processions which rotate through different towns over a 10 to 15 year cycle.

20. Although this was viewed as a static demonstration, in order for the SDL to get to the demonstration, they required a police escort to and from the locality. In essence, this leads to so-called static demonstrations being, in effect, an escorted procession but significantly, falling outwith procedures and legislation governing public processions.

21. While every attempt was made to meet with procession organisers, not all organisers contacted by the research team agreed to meet with them. Requests to meet with the research team were not taken up by organisers from the SDL, Pride of Govan Flute Band and Pride Scotia.

22. See Table AA1, Annex A to see how this varied across the three ward areas.

23. Though hard to quantify only a very small number of business appeared to deliberately shut because of the occurrence of a procession, and in most instances these closures were for a short period of time and were associated with larger processions (notably Coatbridge).

24. These figures are based on notifications and may not, in practice, equate to the actual number of processions that actually took place. See Appendix B for further discussion. They are also based on extrapolated estimates for seven local authority areas where data was not available (see Table AB.1 in Annex B for more details).

25. Based on those local authorities where complete data was available (see Annex B).

26. Separating out events that are clearly political from those that might be said to celebrate 'diversity' (such as events organised by women's groups or the LGBT community) proved conceptually problematic, as even events like Pride Scotia, which may be seen to celebrate a particular identity, can also have a political dimension to them (e.g. securing equality) for at least some participants.

27. The Orr Report referred specifically to Orange Walks; our statistical data includes processions by other loyal orders including the Orange Order.

28. In our national breakdown of local authority areas we were only able to compare areas - and count processions - in terms of individual procession notifications. However, a single notification could relate to what was effectively two processions: an outward procession, typically followed by some form of static event, proceeded later by a return procession that may - or may not - have retraced the original route. Therefore, in counting return procession the figures presented for these three areas here, and in Table AB3 in appendix B, are higher than the figures for the same areas reported in our national breakdown of local authority areas (Tables AB1 in Appendix B).

29. These figures must be viewed as approximate in so far as some events may have been subsequently cancelled without notice. Moreover, in some instances, multiple event notifications may have related to, what in essence, would have been perceived by the public to be one larger event (e.g. a large procession that started with smaller feeder processions). Finally, some processions that may have occurred may have been organised without any notification being given.

30. Figures for procession size are based on details supplied by event organisers in advance. They can only therefore be viewed as rough estimates.

31. The areas will generally be anonymised in the following discussion (from Area A through to D), with local authority and police respondents also being anonymised with numerals (e.g. Police 1, 2, 3 etc. Local authority 1, 2 etc.).

32. This is an extended typology of one provided to us by a respondent in Local Authority 1

33. 'Feeder' and 'return' processions are early morning and late afternoon processions that individual Loyalist lodges or chapters may undertake prior to, and on the return from, joining a larger procession in the middle of the day. Irish Republican processions do not include feeder or return processions.

34. Though if there was substantial evidence of public disorder, intimidation, or criminal damage being likely then the demonstration could be banned from a defined area through a senior police officer making an order under the 1986 Public Order Act. Usually, the available grounds for imposing such an Order have not been considered sufficient.

35. Although Irish Republican processions constitute a very small proportion of all processions.

36. Survey participants were asked for a range of different types of processions that occurred in their community whether they felt 'generally positive, generally negative, or neutral about such events when they happen'

37.The recent Scottish Social Attitudes Survey (2014) provides similar findings to the NFO study. Importantly this report, Community Impact of Public Processions, differs from these studies as our research was conducted in communities that commonly host these processions while the other studies were conducted across Scotland. Another key difference is that the other studies asked questions about these types of processions within the specific context of 'sectarianism' as a social problem. In contrast, our surveys asked more general questions about the positive and negative aspects of all types of processions.

38. See Scottish Neighbourhood Statistics.

39. This inference is supported by a limited amount of post-procession survey work (n=29) that was undertaken in a small sub-area within the centre of Coatbridge that had been over-looked when developing our original sampling frame. This area, around Gartsherrie Road, was distinctly less affluent than our main sampling area and showed far more visible support for the main loyalist procession on the 6th July; and in spite of being no less mixed in terms of religious affiliation/heritage, rated the procession less negatively. For example, 60% of respondents felt that the procession caused tension in the community (compared to 73% across the wider area). 35% felt that the procession caused anti-social behaviour in the community, however 46% strongly disagreed with this statement.

40. Survey respondents were asked for their views on public processions, specifically and generally. This was followed by a series of questions which explored the respondents' community and, in particular, how they felt about it and how safe it felt to live there. The final section of our surveys asked for demographic information which included gender, date of birth, occupation, (dis)ability, country of birth and nationality. Our question on religion asked 'what religion, religious denomination or body - if any - do you belong to?' This was followed by: 'And if you have a religious affiliation, how important is that to you?' with options: very important, important, neither important nor unimportant, unimportant, completely unimportant, don't know/not applicable. This question form and format also applied to other questions: 'Do you have a national identity? (e.g. Scottish, British, Scottish/British, Scots-Irish etc). If yes, could you tell me what it is? (with options denoting importance) and 'Do you consider yourself to belong to a distinct racial or ethnic group? If yes, would you mind telling me what it is? (with options denoting importance).

41. Some 64% per cent of Coatbridge respondents indicated a Catholic background, while only 18% indicated a Protestant background. By contrast, in Parkhead and Govan the distribution of religion was more evenly spread: in Parkhead around a third of respondents were Protestant and just over a quarter, Catholic; in Govan these proportions were almost exactly reversed.

42. Binary logistic regression

43. Measured by a factor score combining responses to survey items such as "I feel like I belong in this area" and "I'm proud to be a resident of [place]".

44. Measured by a factor score combining responses to survey items assessing respondents' views on the extent of racial prejudice toward Black, Asian, Muslims and asylum seekers.

45. For example, 56% of respondents 'strongly agreed' that they felt that they 'belonged' in the area in which they lived both pre and post procession, whilst just under 60% of respondents in both samples rated 'sectarian intimidation or harassment' as being a 'quite' or 'very' common problem in Scottish society.

46. We have no way of knowing in every case whether the procession that respondents were referring to was the same one that we had selected for study.

47. The figure plots results from a simple linear regression model that regressed views of the recent procession (as measured by 5 survey items tapping respondents assessments as to the tension it caused, whether they felt threatened by (a) onlookers or (b) participants, whether they felt angry and whether they felt in danger) on perceptions of sectarianism (as measured by 5 items assessing how common respondents thought sectarian behavior to be). In this model the beta for sectarianism was .60, and the model R2 was .33, indicating a strong correlation between these two variables.

48. Respondents were not asked to provide, or provided with, a definition of 'sectarianism', instead, our surveys were designed using conceptual focusing so at the point where respondents were asked to comment on sectarianism, they were aware of the issues being discussed.

49. Due to the small sample size, and overall low response rates to the postal survey it is not possible to provide further detailed subgroup analysis. Nevertheless these figures give an indication of how perceived experience of sectarian discrimination may influence attitudes to processions. More detailed, national figures on experiences of discrimination can be found in the Scottish Household Survey: . It is worth noting that in our postal survey religious discrimination was more prevalent than discrimination on the basis of ethnic group. However, this is likely to be due to the particular characteristics of the case-study areas and the small number of ethnic minority members surveyed (for more details see Annex A).

50. These percentages exclude the small number of 'don't knows' that represented between one to 8% of all responses. The item generating the most 'don't knows' was 'freedom of speech'.

51. Two different questions were used to try and pinpoint a respondent's relationship to a procession. In most instances the relationship was fairly clear-cut, but in some instances a respondent may have characterised themselves as bystander, but then having 'heard' or 'come across' the procession, the degree of their involvement (in terms of their engagement, or the duration of their involvement) may have arguably shifted their status to that of participant.

52. Referred to throughout this report as bystanders.

53. Totals per question vary as they omit 'don't know' responses as well as any refusals to answer the question.

54. Though businesses did not have to be actually on the procession route, businesses that were within site of the route, or which were close enough to be clearly potentially impacted by the presence of the procession, were also targeted.

55. However, in circumstances where we knew businesses had deliberately 'closed shop' for the duration of a particular procession (and this was uncommon outside of the largest procession site), we attempted to contact and survey those business owners later.

56. Though some respondents talked about closing businesses for larger processions, more common forms of mitigating actions would include removing stock from outside, or ensuring that staff arrived earlier than usual to avoid being caught in any procession-related traffic restrictions.

57. Note that this list does not exhaust the types of incidents included in the police database, and a range of other events or crime types are excluded. These are generally either very rare or serious incidents (e.g. abduction/extortion), or, more often, incidents prima facie unconnected to the issue at hand (ranging from airport emergencies to untaxed vehicles).

58. Note that two marches occurred in Coatbridge, the Grand Lodge procession on 6 July and the annual Cairde na h'Eireann procession on 13 July.

59. Regardless of whether they were there to deliberately watch a procession or not.

60. Although both the SDL and UAF were excluded from Pollokshields following the events of January 2013 under Section 14 (2) (b) of the Public Order Act 1986.

61. Assumed to refer to the Provisional Irish National Liberation Army

62. Though it needs to be recognised that in other instances, where there is information to suggest that a procession may attract a significant risk of disorder, a 'light touch' policing approach will not be possible.

63. See Annex F for indications of the cumulative impact on one geographical area.

64. Due to limitations of size we have not included research tools in this Annex. If you would like to obtain a copy of surveys or questionnaires, please contact the research team at the University of Stirling.

65. Structured observations have been used in previous research (for example in Euro 2000 and in UEFA matches in 2004).

66. Though our original research design had made some focus of the pre-post element to the survey as a way of measuring whether a residents' exposure to a procession made a difference to how they felt about their area (in terms of crime, anti-social behaviour, sectarianism, community strength etc.), the low number of post-telephone responses did not facilitate a meaningful pre-post analysis. However, given our subsequent experience of working in these areas, this element now seems misguided. The communities we surveyed, with one exception, hosted multiple processions all summer, every summer. The notion therefore that one more procession should somehow fundamentally change a residents' opinion, may have been therefore in retrospect, naïve.

67. Notification estimates are provided by the procession organisers and do not necessarily equate with the number of participants who attend. Moreover, one notable distinction between Loyalist and Irish Republican events is that Loyalist notifications provide estimates for members of the loyal orders (i.e. members of the organisation) while Irish Republican processions often include cummain members and supporters, who are all encouraged to join the procession. This means they can be stewarded by the organisation unlike Loyalist supporters who often remain outwith the procession and therefore, in the past, outwith the control of official stewards.

68. χ2=8.34, p=.02

69. χ2=8.89, p=.01

70. The figure plots estimates derived from a binary logistic regression model predicting the probability of holding negative views. Age and a quadratic effect for age were included to capture the non-linear effect shown. The shaded area around the line shows the 95 per cent confidence interval.

71. Measured by a factor score combining responses to survey items such as "I feel like I belong in this area" and "I'm proud to be a resident of [place]".

72. Produced in the same way as Figure AC.1.

73. Measured by a factor score combining responses to survey items assessing respondents' views on the extent of racial prejudice toward Black, Asian, Muslims and asylum seekers.

74.The figure plots results from a simple linear regression model that regressed views of the recent procession (as measured by 5 survey items tapping respondents assessments as to the tension the procession caused, whether they felt threatened by (a) onlookers or (b) participants, whether they felt angry and whether they felt in danger) on perceptions of sectarianism (as measured by 5 items assessing how common respondents thought sectarian behaviour to be). In this model the beta for sectarianism was .60, and the model R2 was .33, indicating a strong correlation between these two variables.


Email: Linzie Liddell

Back to top