by Claude Opus 5.5
Armstrong, Hamilton and Sweeney explain party rape through interacting individual, organisational and cultural factors. Which level is most tractable for ethical intervention, and which most morally salient?
In a widely cited 2006 article in Social Problems, the sociologists Elizabeth Armstrong, Laura Hamilton and Brian Sweeney drew on an ethnographic study of women on a residence hall floor at a large American university to explain what they called party rape: sexual assault that occurs at or after parties, typically involving alcohol and acquaintances. Their central claim was that party rape is not best explained by individual pathology alone. It is the predictable product of interacting factors at three levels: individual characteristics and expectations, the organisation of social life on campus, and the interactional scripts and cultural assumptions that govern partying. The case behind these seminars, which involves allegations about events at a fraternity house during a night of drinking, sits squarely within the territory they described. Their framework invites a question about where ethical effort should go: the level where intervention is most likely to work, or the level where moral responsibility seems most concentrated. These may not be the same.
The three levels
At the individual level, the authors described students arriving at university with expectations of fun, social status and sexual experience, and with gendered beliefs about how men and women should behave. Many first-year women, eager to fit in, sought out parties as the centre of social life.
At the organisational level, they emphasised how campus life is structured. Residence halls at the university they studied were alcohol-free and closely policed, which pushed first-year students towards fraternity parties, where alcohol was available. Fraternities controlled those parties: they owned the space, supplied the alcohol, decided who entered and, often, provided transport. Men were hosts; women were guests. The university’s own policies thus helped to channel young women into spaces controlled by older men.
At the interactional and cultural level, they described scripts that made assault more likely and harder to name: norms that women should be friendly and trusting towards hosts, that heavy drinking was part of the fun, and that women who were sexually active risked being judged. These norms discouraged women from leaving uncomfortable situations and from reporting what happened afterwards.
Tractability
If the question is where intervention can most reliably change outcomes, the organisational level is the strongest candidate. Organisations have levers. Universities set residence policies, recognise or derecognise societies, control funding, own or regulate property and can require hosting standards. A university can change who controls party spaces, require events in university-managed venues, or reconsider policies that make fraternity houses the main place where first-year students can drink. These changes do not require changing anyone’s beliefs; they change the setting in which beliefs are acted on.
Individual-level interventions have a more mixed record. Consent education is valuable but evidence about its effect on behaviour is uneven. Bystander programmes appear more promising, partly because they work at the boundary between individuals and the group. One of the most striking findings comes from a randomised trial, published in 2015 by Charlene Senn and colleagues, of a resistance programme for first-year women in Canada, which substantially reduced reported rape over the following year. That is a significant result, but it illustrates an ethical tension: interventions that work by asking potential victims to protect themselves locate the burden in the wrong place.
Cultural change is the least tractable in the short term. Norms about masculinity, alcohol and sex are shaped far beyond the campus and are slow to shift, though organisational change can over time alter what is normal within a community.
Moral salience
If the question is where moral responsibility is most concentrated, the answer looks different. The person who commits an assault is the most direct and culpable agent. No account of organisational or cultural causes alters that. Indeed, one risk of multi-level explanation is that it can seem to dilute individual culpability by presenting assault as something that just happens in certain settings.
Yet the organisational level is also morally salient, in a different way. Armstrong, Hamilton and Sweeney’s account shows that institutions designed environments in which assault was foreseeable, and did so while knowing the risks. Moral responsibility tracks not only causal proximity but also knowledge and the capacity to prevent harm. A university that keeps in place policies it knows funnel students into high-risk settings bears responsibility for that choice, even though it commits no assault.
The cultural level is morally salient in a diffuse way. Scripts and norms are maintained by countless people, including those who would never commit an assault but who laugh at the jokes, judge women who report and treat heavy drinking and sexual conquest as markers of status. Responsibility here is real but widely shared, which makes it easy for everyone to disown.
When tractability and salience come apart
The mismatch matters. If we concentrate only on the most culpable agents, we focus on punishment after harm has occurred. If we concentrate only on what is tractable, we may adopt measures that work but that implicitly treat assault as a risk to be managed rather than a wrong to be condemned, and that may load the burden onto potential victims.
A defensible approach holds both in view. Organisational reform should be pursued because it is effective and because the institutions that control campus space are responsible for the risks they create. Individual accountability should be maintained because assault is a wrong done by a person, not by a system. And cultural work should continue, not because it will produce rapid change but because a community that does not contest the norms that make party rape possible cannot claim to be doing all it can. Armstrong, Hamilton and Sweeney’s framework is useful precisely because it refuses to let any one level carry the whole explanation, and therefore the whole responsibility.