Racism and the criminal justice system in the US

(1) Looking through the Washington Post data on police shootings, a few things jump out.

Police killings are quite rare in the US. Since 2015, according to the Washington Post, 2,490 people have been killed by the police. 7% of these, 174, have been unarmed. Out of that 174, 65 were black, 67 white and 35 Hispanic. Out of the entirety of police shootings, armed or unarmed, 27% were black, 18% Hispanic, 52% white. There’s no breakdown by income, though geographically a significant proportion of the shootings(32%) occurred in three states, California, Florida and Texas

(2) The expansion of the US prison population is not primarily driven by the war on drugs. Most people, outside of federal jails, are incarcerated for violent crimes, not non-violent drug offences:

The core failing of the Standard Story is that it consistently puts the spotlight on statistics and events that are shocking but, in the grand scheme of things, not truly important for solving the problems we face. As a result, it gives too little attention to the more mundane-sounding yet far more influential causes of prison growth. For example, a core claim of the Story, made perhaps most forcefully by Michelle Alexander in her book The New Jim Crow: Mass Incarceration in the Age of Color-Blindness, is that our decision to lock up innumerable low-level drug offenders through the “war on drugs” is primarily responsible for driving up our prison populations. In reality, only about 16 percent of state prisoners are serving time on drug charges—and very few of them, perhaps only around 5 or 6 percent of that group, are both low level and nonviolent. At the same time, more than half of all people in state prisons have been convicted of a violent crime. A strategy based on decriminalizing drugs will thus disappoint—and disappoint significantly. Yet we see little to no efforts to reform the treatment of people convicted of violent crimes. The Standard Story also argues that increasingly long prison sentences have driven growth, and thus that cutting back sentences would effectively cut prison populations.. The far more significant change, as I will explain more fully throughout this book, is the increased rate at which people get sent to prison in the first place. The primary driver of incarceration is increased prosecutorial toughness when it comes to charging people, not longer sentences. Stopping prosecutors from sending people to prison to start with would be far more effective in cutting incarceration rates than reducing the amount of time prisoners spend in prison once they get there—and this fact points to a very different set of reforms than those generally proposed.

There’s evidence of bias in sentencing between whites and African Americans, but the disproportionate incarceration rate is driven by the fact that African Americans commit more serious crimes, more often:

Despite the persistent statistical and measurement obstacles noted by the NRC, Baumer, and other experts, we still know more than we think we know today about racial disparities in punishment. We now know that the significance of racial bias and race of the defendant in fostering sentencing disparities was grossly overstated, especially for serious and violent crimes. Comprehensive reviews of the research on racial disparities in sentencing have found that race of the defendant has only a relatively weak influence on sentence length. As for the decision on whether to sentence someone to prison or impose an alternative sanction, the defendant’s race has a significant but relatively modest effect overall and is highly variable across studies. We also know that race effects are highly contingent on many other factors, including age, gender, socioeconomic and employment status, quality of legal representation, offense type, criminal record, and jurisdictional location. We now have a more subtle understanding of how racial discrimination and bias operate in the criminal justice system. Recent research strongly suggests that the flagrant Scottsboro-type discrimination is not a primary driver of racial disparities in criminal justice outcomes. Rather, it appears that a little bias can still go a long way in a system that has many decision-makers—police who must decide whether to make an arrest or not; prosecutors who determine whether to dismiss the case, what to charge, and what plea bargains to seek; and judges who determine bail and sanctions.


Although the persistence of racial biases and discrimination in the criminal justice system is deeply troubling, we now know these are not the only reasons why African Americans are disproportionately incarcerated. Two other key factors are that blacks disproportionately commit the types of crimes that usually draw a long prison sentence, and that the sentences and time served for those offenses have escalated since the 1970s. Cassia Spohn concludes that the evidence is “irrefutable” that seriousness of the offense and the offender’s prior record are the primary determinants of who goes to prison and for how long.

(3) Read Rajiv Sethi respond to Roland Fryers paper that argues:

Given the stream of video “evidence”, which many take to be indicative of structural racism in police departments across America, the ensuing and understandable outrage in black communities across America, and the results from our previous analysis of non-lethal uses of force, the results displayed in Table 5 are startling… Blacks are 23.8 percent less likely to be shot by police, relative to whites.

I think three things seem to be true. (1) There’s evidence of discrimination and bias against African Americans, to varying degrees of importance, across all socio-economic levels in the US. (2) Low income African Americans face numerous structural barriers to education and employment that seriously undermine any possibility of social mobility, and incentivise criminal behaviour . (3) The public debate on racism in the criminal justice system is mostly wrongheaded.


The dead of the Irish Revolution

From Eunan O’Halpin‘s project ‘counting the dead of the Irish Revolution’, and chapter 8 in the book Terror in Ireland :

                           Between January 1917 and December 1921, 2141 fatalities have been recorded. That number may be broken down by the historic county (or country outside Ireland) where each death occurred:


Cork 495                             Wexford 23                Dublin 309          King’s Co. (Offaly) 21

Antrim 224                                Donegal 20                         Tipperary 152       Kilkenny 19

Kerry 136                                 Westmeath 18                         Limerick 121         Sligo 18

Clare   95                                  Meath 17                         Galway   58             Tyrone 16

Roscommon   58                           Leitrim 15                         Mayo   43     Carlow 13

Londonderry   41                    Kildare 12                         Waterford   35     Great Britain 11

Armagh   28                      Queen’s Co. (Laois) 10                     Down   28            Cavan 9

Longford   26                             Fermanagh   9                    Louth   26            Wicklow 7

Monaghan   25                                 India   3

The number of fatalities in each county is, of course, a very crude index of the intensity of disruption experienced by people and communities during the Irish revolution…..Assigning responsibility for the 2141 fatalities reported is not a matter of simply splitting the total between rebels and government forces. The IRA were definitely responsible for 46% of these fatalities, and Crown forces for 42%. A further 6% were the work of undenominated snipers or rioters, almost all in sectarian clashes in Belfast. 2% occurred in cross-fire between Crown forces and separatists, 1% were killed by loyalists, 1% by civilians, and the killers of the remaining 2% are unknown Within the aggregate of 2141 deaths, 898 (48%) were civilian and 1243 (52%) combatant.

The civilian proportion rose from 39% in 1920 to 45% in 1921. This confirms Peter Hart’s finding that the proportion of civilian fatalities grew in 1921, which was also the most violent year accounting for 61% of all deaths between 1917 and 1921. However, Hart’s estimate that 64% of deaths in 1921 were civilian is clearly excessive. The discrepancy cannot be explained simply by methodological differences, such as our inclusion of deaths in motor traffic accidents involving police and military vehicles, which some might find problematic. Rather, it arises mainly from the restricted range of sources available when Hart conducted his research. Of the 1243 combatant fatalities, 467 (38%) were IRA, 514 (41%) police and 262 (21%) British military. Within these categories, 9% of IRA fatalities were inflicted by the victim or his comrades, as were 14% of police fatalities and no less than 26% of British military fatalities. Accidental shootings were mainly responsible, along with premature explosions in the case of the IRA and motor vehicle accidents in the case of the police.

It seems remarkable that a professional army should lose so many men in this manner, particularly by comparison with the hastily expanded police and the part-time, under-trained IRA…. Determining responsibility for the 898 civilian fatalities is quite challenging, particularly in Antrim. For the 194 civilians killed there during riots and sectarian clashes, it is often impossible to determine who fired a particular shot, and we can be fairly sure of responsibility in only 91 cases. Across the country, many civilians died at the hands of mixed groups of soldiers and police. Consequently, those forces have been banded together in assigning responsibility. The IRA may confidently be held responsible for 281 (31%) of civilian fatalities, compared with 381 (42%) attributable to Crown forces. A miscellany of rather vague categories (loyalists, snipers, civilians, either IRA or Crown forces) accounts for the remaining 236 (27%).

Only 92 (10%) of all civilian fatalities were female. The majority of these were untargeted killings in riots, or traffic accidents involving Crown forces. The IRA abducted and shot at least three women as spies in 1921 (Bridget Noble and Mary Lindsay in Cork, and Kate Carroll in Monaghan). Two of the victims had a reputation for social deviance: Kate Carroll kept an illicit still and Bridget Noble, although married, was held to be too friendly with a number of RIC men on the Beara peninsula. These killings caused acute embarrassment locally and at GHQ.

The project elucidates another vexed question, the experience of ex-servicemen during the War of Independence. Between 1919 and 1921, 420 ex-servicemen were killed, constituting 19.6% of all deaths. Of these, 227 (54%) were serving policemen, so their fates cannot be ascribed to their previous military service. Of the remaining 193 cases, 16 (4%) died as IRA men, leaving 177 (42%) who were civilians. Of these, the IRA clearly killed 99 (56%) and Crown forces 46 (26%). In the remaining 32 cases, including 13 in Antrim, it is impossible to determine responsibility.

What is more telling is the distribution of responsibility for civilian ex-servicemen’s deaths across the country. In Antrim, 26 civilian ex-servicemen were killed, mainly during riots or sniping in Belfast. In all 13 cases where responsibility can reliably be assigned, Crown forces were to blame. In Dublin, the IRA killed 14 of the 25 civilian ex-servicemen who died. Even there, it is dangerous to assume that all were singled out on account of their backgrounds: for example Frank Davis, caretaker in the Custom House, was shot because he supposedly went towards a telephone, not because he was a Boer War veteran. In Cork, the most violent county, 49 civilian ex-servicemen were killed. Of these, 32 were definitely victims of the IRA, while Crown forces killed 15, including one shot when caught in the act of raping a young girl (meaning that Cork accounted for almost a third of all civilian ex-servicemen killed by the IRA and likewise by Crown forces).

Though the evidence that ex-servicemen were systematically targeted by the IRA is strongest in Cork, more than half of all civilian casualties in two relatively ‘quiet’ counties were ex-servicemen, all killed by the IRA. In King’s (Offaly), the figure was 6 (60%), and in Meath 5 (50%). The 35 fatalities occurring on Bloody Sunday reflect the diversity of circumstances, often contested, associated with revolutionary killings in general. As noted above, members of the IRA (3), Crown forces (16) and civilians (16) were all represented, the civilian proportion approximating that for all fatalities (48%). All died after being shot, the predominant means by which death was inflicted throughout the War of Independence. All fatalities were investigated in an expeditious but cursory manner through military courts of inquiry. What sets Bloody Sunday apart is the three spectacular events for which it became notorious: the co-ordinated assassinations in the morning, the indiscriminate shooting of civilian spectators in the afternoon, and the killings in custody of three prisoners during the night. None of these three episodes, however, was truly representative of the conduct of either the IRA or its enemies during the Irish War of Independence. To that extent, the events of Bloody Sunday were sui generis.