Representations of Abuse: What are they and why do they matter?

Rose coloured glasses (Image from Pixabay: By MabelAmber Free to use or share).

Domestic or partner abuse is a pervasive problem in our society, and in Australia is almost universally recognised as such. Although there is a near consensus acknowledgement that domestic abuse is harmful and should not be tolerated, there seems to be less consensus on what people see, or don’t see as abuse. Given that challenging abusive behaviours first requires the recognition of abuse, this highlights a need to understand how people’s concepts of abuse might help or hinder their identification of abuse in intimate partner relationships.

Prior research has shown that people are uncertain about whether non-physically abusive behaviours are abuse. Most would agree that a person who uses a weapon against or beats their partner is guilty of abuse, but it is well known that in many relationships in which abuse occurs the individual behaviours do not reflect these overt forms of physical abuse. In fact, many abusive relationships are characterised by harmful behaviours that are non-physical in nature (e.g., accusations of flirting). This is particularly true in the early stages of an abusive relationship where subtle, non-physically abusive behaviours are typical. However, it is important to note that whilst individual behaviours may appear relatively minor, this does not negate the considerable harm these behaviours can cause. This is largely due to a key discrepancy between abuse and other crimes. Domestic abuse differs from many other crimes (e.g., theft) in that the harm is cumulative, caused by multiple incidents over an extended period, rather than a single incident at a single point in time. Consequently, the harm caused by each incident builds on the harm caused by each prior incident.

In my research I wanted to explore people’s concepts of what partner abuse looks like to gain insight into how they might perceive early abusive behaviours. Specifically, I wished to determine the types of behaviours they might expect to see together, and whether previously observed differences in judgements of non-physically and physically abusive behaviours persist when people are presented with multiple behaviours rather than a single incident.

What we did?
In our study, participants were presented with two non-abusive behaviour exemplars (A agrees to try B’s solution [to an argument or problem]; A shows they care about B), two non-physically abusive exemplars (A accuses B of flirting; A shouts at B), and two physically abusive exemplars (A grabs B; A slaps B). For each exemplar behaviour, participants were also presented with a list of behaviours which depicted non-abusive, non-physically abusive, and physically abusive behaviours. They were then asked to select up to four behaviours they thought likely to occur in a relationship where the exemplar behaviour was present. After creating each of these six behavioural clusters, participants were asked to indicate whether they would judge a relationship where the exemplar and their clustered behaviours were present to be abusive, possibly abusive, or not abusive.

What we found?
Broadly our results indicated that participants expected similar types of behaviours to cooccur. Participants were most likely to select non-abusive behaviours to cluster with non-abusive exemplars, non-physically abusive behaviours to cluster with non-physically abusive exemplars, and physically abusive behaviours to cluster with physically abusive exemplars. Even when judging clusters with multiple abusive behaviours, participants were less certain in the identification of abuse when presented with clusters of non-physically abusive behaviours than when presented with clusters of physically abusive behaviours.

An additional interesting finding was that although non-physically abusive behaviours were most likely to be chosen for both non-physically abusive exemplars, there were distinct themes present in the two non-physically abusive exemplar clusters. The behaviours most frequently clustered with the flirting exemplar seemed to reflect a theme of controlling and monitoring behaviours, with non-abusive and physically abusive behaviours similarly unlikely to be selected. In contrast, the behaviours most frequently clustered with the shouting exemplar reflected a theme of toxic conflict with physically abusive behaviours more likely to be selected than non-abusive behaviours. Participants viewed the flirting exemplar clusters as less clear evidence of abuse than the toxic conflict clusters.

What might it mean?
Broadly these results suggest that participants expect similar types of behaviours to cooccur.  Interpreted in light of the abusiveness ratings, the results also suggest a hesitancy or uncertainty in the identification of non-physically abusive behaviours as abuse, even when multiple behaviours are presented simultaneously. This was particularly true of controlling behaviours linked to the flirting accusation exemplar. Given that early abusive behaviours are typically non-physical, these findings suggest efforts to combat the occurrence of domestic abuse in the early stages may be undermined by a persistent failure to recognise non-physically abusive behaviours as abuse, a failure that is particularly pronounced when the behaviours align with narratives of jealousy.   

 - By Kiara Minto

For Further Reading
Minto, K., Masser, B., & Louis, W. (2021). Lay Understandings of the Structure of Intimate Partner Violence in Relationships: An Analysis of Behavioral Clustering Patterns. Journal of Interpersonal Violence, 088626052098627. https://doi.org/10.1177/0886260520986276