Background to the Research

• This research set out to examine the ethnic memory and ethnic identity of two cohorts of Northern Irish students, in order to explore the possible relationship between memory and social identity in the context of a long-standing ethnic conflict.
  

Research Approach

• In 1984 and 1995, university students were asked to classify themselves in terms of being British or Irish. They were shown 12 video clips of political events in Northern Ireland from 1969 to 1980 and asked to date each clip. Then they indicated on a scale how that particular event made them feel in terms of being Irish or British.
  
  
• In 1984 all the respondents were university students (n=100) with a mean age of 20.13 years. There were approximately equal numbers of Catholics and Protestants. In 1995, the sample again comprised of all university students (n=115) and the mean age 24.16 years. The majority of participants in this sample were Catholics (70%).

Main Findings

Baseline Identity

• All participants rated themselves on a 5-point scale from 1 (very Irish) to 5 (very British) to establish their 'baseline' identity. The mean baseline identity for the 1984 sample was 2.18 for Catholics and 3.60 for Protestants, indicating that Catholics tended to rates themselves as 'Irish' and Protestants tended to rate themselves as 'British'.
  
  
• Similarly in the 1995 sample, the mean baseline identity was 2.04 for Catholics and 3.50 for Protestants.

Identity in Response to Events

• For the 1984 sample, the analysis of responses to video events showed that there was a main effect for religion and event classification. It also revealed statistically significant interaction between participant's denomination and event classification.
  
  
• The main effect was due to Catholics recording a more Irish identification in response to all events than did Protestants, whilst the main effect for event classification was due to Nationalist events eliciting a slightly more Irish feeling than did Unionist events for all participants.
  
  
• Inspection of relevant means would suggest that the Event X Participant's Denomination interaction was due to Catholic participants reacting in a slightly less Irish way to Unionist events than they did to Nationalist events. In contrast, Protestant participants reacted in apparently a uniform (largely British) manner to both Nationalist and Unionist events.
  
  
• For the 1995 sample, the analysis indicated that there was a main effect for religion and for event classification, however no statistically significant interaction between participant's denomination and event classification was found.
  
  
• The main effect for religious denomination was due to Catholics recording a more Irish identification in response to all events than did Protestants. The main effect for event classification was due to Nationalist events eliciting a slightly more Irish feeling than Unionist events for all participants regardless of denomination.

Memory Test

• The mean number of correct answers on the memory test in the 1984 sample indicated overall performance level at well above chance.
  
  
• There was a statistically significant interaction between participant's denomination and event classification and a main effect for event classification, due to Nationalist events being dated more accurately than Unionist events.
  
  
• For all participants in the 1995 sample, the mean number of correct answers on the memory test indicated an overall performance at above-chance level.
  
  
• Analysis of the 1995 responses also indicated that there was no effect for religion but that there was a main effect for event classification. There was a statistically significant interaction between participant's denomination and event classification, due to Nationalist events being dated more accurately than Unionist events.

Relation of Identity to Memory

• In both the study years, 1984 and 1995, there were no statistically significant relations between baseline identity and memory test performance. However, in 1984 and 1995 for Catholic participants only, there were significant negative correlations between variability and memory test performance. These findings suggest that Catholics who tended to vary across situations and thus had high variability scores tended to know less about the 'troubles'. In neither year did this relation emerge for Protestant participants.

Conclusions

• There were virtually no differences between both groups in terms of their baseline identity. Despite an eleven year gap, in both samples, the Catholic participants felt equally Irish and the Protestant participants felt equally British.
  
  
• Events presented on video did influence the strength of identity. In particular, in both years, Catholics responded in a significantly less Irish way to Unionist events than they did to Nationalist events.
  
  
• Only in 1984 was there evidence of clear superiority for Catholics over Protestants when asked to date Nationalist events. In fact the data indicated that Nationalist events were dated best in both years, and further by both Catholic and Protestant participants.
  
  
• Although participants may not always have remembered exactly when events took place, this did not prevent these events from 'switching on' the relevant social identity. This may suggest that in Northern Ireland and perhaps in other societies such as Bosnia and the Middle East, ethnic conflict may not rest entirely on ancient quarrels and present-day events but may be being reinforced by events from the more recent past.