S11, E7: Rekindling Old Flames
S11, E7: Rekindling Old Flames
Series 11: The Collateral Damage of Revenge in Science
Previously in
it was mentioned that the American Astronomy Association (AAS) and its Committee for the Status of Women in Astronomy (CSWA) were seriously discussing the idea of academic shunning as a retaliation against suspected sexual harassers.
Blogs on the Women in Astronomy website seemed to have reached an odd consensus on due process i.e., it was a failed concept that should be bypassed, rather than one that is inconsistently applied and in need of rehabilitation. If you were to sustain a knee injury, you wouldn’t expect the physician or physio to tell you, ‘it was caused by having legs’.
The inspiration for this was an article by three philosophers whom, starting from the reasonable premise that certain complaint processes were often ineffective, reframed the issue to make the requirement for evidence the problem. It’s often a disregard for the available evidence that compounds the threat to female safety. How can ignoring evidence be both the problem and the solution - is it meant to be homoeopathic?
That any scientist would be convinced that this had merit is surprising enough, but for scientific organisations to think it is worth considering is in the realms of the ridiculous. Rather than seeking better complaint processes to improve female safety, reluctant women were conscripted into taking down a celebrity scientist, as a political flex. In the case of Geoffrey Marcy, any woman who may have had contact with Marcy at Berkeley in the past were approached for recollections of any interactions that might be scrutinised.
We will come back to all that but in this episode I am going to take a diversion to be constructive about some of the fundamental issues. First to kick off a new safety engineering theme (that will be embedded into the next few episodes) before looking at the end-to-end effectiveness of legal processes for rape in the United States.
Safety Engineering - Part 1: The Safety Pyramid
The safety pyramid/triangle is a visual representation of the relationship between levels of incident severity and the ratio of escalation (or probability). Typically, this would be devised empirically on major worksites. In the hypothetical example illustrated, for every 300,000 ‘At-Risk’ behaviours there are 3,000 ‘Near Misses’; 300 Recordable Injuries’ (often called Loss Time Injuries (LTI); 30 Lost Workday Cases and 1 one fatality. The broader the base greater the number of incidents at every level.
At the lowest level the behaviours might result in poor site housekeeping, such as running power cables across walkways, leaving equipment in places where it becomes a trip hazard or a failure to wear suitable PPE. Therefore, by tackling the low level behaviours the number of incidents and escalations are reduced, made achievable by making everyone responsible for their actions.
There are two possible undesirable consequences of this model depending on how it is used and I have seen many examples of both over decades, around the world, both onshore and offshore.
Under reporting: this happens where the emphasis is placed on ‘days without an LTI’ which incentivises the hiding of incidents, because there is group pressure on anyone who has been injured to hide it. I was once on a ship in Nigerian waters and one of the crew broke his leg, so the offshore manager put him on ‘light duties’, so it wouldn’t be logged as an LTI.
Over reporting: this happens when reporting ‘near misses’ is incentivised by rewards such as electronic goods. This can also lead to vexatious and retaliatory reporting.
To create a good safety culture it is important that incidents can be accurately monitored and free of skewing from misguided application of incentives. Incrementally improving the process requires consistency with small adjustments. Gaming the process rather than optimising it will impact safety overall.
These are elements that you need to understand what I say about improving female safety later in the series. In essence this means ensuring integrity and insisting that even the pseudo-legal processes are effective. Without respect for the process there is no process; if due diligence is not held to be sacred there can be no justice either.
So what do the statistics tell us about sex crime in the United States?
Calibrating for California
Let’s look at some statistics for rape in the USA in one year. Below is a snapshot of an interactive table (the actual date range is not specified but a query has been submitted to Wisevoter1). In the excerpts below I have filtered by the number of reported rape incidents. The original table can be found at the following link: Rape Statistics by State.
In Illinois the likelihood that a reported rape will lead to an arrest is 0.3%, the lowest in the entire country. The justice system there provides the least incentive to report rape, but it is still #7 out of 50 States in terms of reports, so what might that tell us about the relationship between legal process and crime?
California is the most populous state which is the biggest factor contributing to all its crime rates, but if the arrest performance of Illinois was applied to it, there would only be 56 rape arrests rather than 2060. It’s a bit of an over-simplified take but it is not unreasonable.
At the other end of the table we have this2.
In Vermont a reported rape has a 22% chance of resulting in an arrest. Apply that performance to California and there would be 3986 arrests. I suggest these disparities show the application of law across the United States, at least as it applies to sexual violence, to be woefully uneven. This cannot be justice.
Undoubtedly the differences in the rates of sex crime, reporting, arrest and conviction across the states are cultural, and they confound each other, so for example, if the chances of arrest and conviction are low, crime is incentivised as reporting is disincentivised, which creates a less safe environment for women.
Minnesota Legal Transition Rates
What about the performance of the legal process from incident, report, arrest, prosecution, conviction and incarceration? Let’s call these waypoints ‘states’ and the probability of transitioning from one state to the next the transition rate. Accordingly the site gives the national averages as follows:
2,959 Reported Rape Incidents
331 Rape Arrests
501 Sex Offenses Arrests
Minnesota is at #17 in the table but its reported rapes are almost bang on the national average, viz.
I am using Minnesota’s figures because I was able to find the statistics on this state quite easily. Below is a table of ‘legal transition rates’ that I have compiled from information found from the Central MN Sexual Assault Center.
2012/3 marked the beginning of a sharp increase in the number of rapes nationally. Mightn’t this explain the elevation of alarm on campuses, the sense of urgency and the wiliness to subscribe to any initiative to fight back, by whatever means?
Perhaps the justice system in the US is so capricious we cannot really know what the true rates of sexual crime are. Without a federal recalibration (which is unlikely to be a realistic aspiration) and in the absence of a better idea, it seems inevitable that universities will foster a vigilante culture. In consequence, faculties are already willingly handing power to administrators, because dealing with these issues are too politically and personally risky. Under those conditions where retribution is the main objective, how can Title IX be a just process, not just for everybody but for anybody?
Next: S11,E8: The Preemptive Strike
The B-Side
From March 26 I will switch on paid subscriptions. After that, on selected series I will provide some paid companion ‘episodes’ (my work is typically episodic). These will generally complement the free episode without being essential to it. If this is unclear or you are curious about what this is about please see ‘The B-Side Explainer’.
The link below is to a free B-Side.
This just for good housekeeping because in every other respect, it should make no difference, given I am using the figures to compare performance between states.
Note that the 51 States listed include District of Columbia (DC) which of course isn’t a State but a federal city that is accountable to US Congress and is otherwise beyond the governance of any State.