The Call Is Coming 3: A Forman Moment for Arab Israelis? And Why Expect So Little from Your Taxes?

In the previous two installments of this series, I discussed parallel processes I see in Israel and in California: rising crime rates and resulting miseries within underserved communities–Arab-Israeli towns and villages, and Black communities in Bay Area cities (disproportionately affecting these communities both in terms of perpetrators and victims). In the first installment, I showed that these issues have yielded calls from “inside the house” to improve police response to crime rates. In the second installment, I discussed a curious difference: the Arab Israeli calls are monolithic and they demand solidarity from allies in securing police presence and protection, whereas Black American calls–the recent NAACP letters in Oakland and San Francisco–are heavily contested and far from representative of the defund-abolish-dismantle-repeal sentiment. I proposed a few differences between the two scenarios and concluded that the problem is one of intra-movement politics.

In this last installment in the series, I want to posit two additional issues: one of timing and one of civic expectation. The first is, in some ways, a continuation of the argument I made in my article Bad Role Models, in which I discussed American influence on Israeli criminal justice. In that article, I showed how criminal justice developments in the U.S. migrated to Israel through a process of elite networking, often with a delay of 15-20 years, to the point that Israel implements American policies long after empirical evidence already undermines their merit or efficacy. I listed four developmental stage: the rise of American criminal justice as a model of influence; the “decade of rights”, inspired by the mistaken perception in 1990s Israel that American criminal justice is pro-defendant; the “law and order period” in the 2000s, in which Israel adopted victim rights and anti-sex-offender paradigms that were already being eschewed in the U.S.; and the “era of contrition”, in which new Israeli elites, who learned about mass incarceration in the U.S., started chipping at the punitive block.

My friend and colleague Hagit Lernau thinks that the Arab Israeli faith in policing as an answer to violent crime might be temporary, an echo of the period in the 1980s and 1990s in which Black politicians and police chiefs in D.C. wanted massive police intervention in the crack epidemic. In Locking Up Our Own, James Forman found great empathy for these Black power brokers, even though, as an abolitionist, he disagrees with them. He does not think the crime problem was exaggerated or did not exist–he fully admits that the calls for more policing came out of real distress that was grounded in fact–even as he rejects the premise that aggressive enforcement could have improved things.

To understand Hagit’s argument, let’s locate Forman’s politicians and cops along a timeline. Their preoccupation with internal community problems of crime can be seen as a retreat from Martin Luther King Jr.’s general message of a great project of equality, as well as from Malcolm X’s general message of militant opposition to white supremacy, toward sectorial interests of personal safety within Black cities and neighborhoods. This retreat, which happened in the 1980s-1990s, can be seen as a harbinger of the Arab-Israeli retreat from full commitment to the idea of Palestinian liberation/independence toward sectorial interests of citizens within the 1967 borders. If so, we might expect that the later developments in critical race perspectives on criminal justice–the academic concerns about police oppression and race and their migration to the mainstream of the progressive movement–might eventually make it in Arab-Israeli societies, perhaps through a process of elite networking (or through some other process) and we simply have to work through the delay. But eventually the moment of yearning for police will pass, and we’ll be in a defund/dismantle/abolish/repeal moment in Israel, too.

Here’s another theory on how this could happen: Perhaps, as in the case of D.C., the disillusionment that accompanies massive, oppressive police presence will cool the population’s enthusiasm for enforcement. A couple of weeks ago I talked to a friend who is a police detective investigating serious crimes, including in Arab-Israeli towns and villages. My friend tells me that, as soon as a serious crime is committed in a village, the police’s modus operandi is to send in border patrol officers, who proceed to harass and humiliate everyone around them and make life in the village unbearable. Unsurprisingly, after a few weeks of this, the officers who want to actually solve the crime encounter a wall of silence and mistrust. It is only a question of time until this realization becomes generalized and the community nationwide will stop calling for the police to help.

Which brings me to my second point, the issue of civic expectation. The famous serenity prayer invites us to have the wisdom to tell apart things that can be changed (and require courage) from things that are immutable (and require serenity.) The Forman moment, as well as the current moment in the Arab-Israeli crime prevention movement, assume that crime-ridden streets can be cleaned and that the erosion in public safety can be stopped, or even reversed, if the Israeli government wakes up from its appalling neglect and acts. The Defund movement makes the opposite assumption: nothing good can come from police intervention, so they might as well stay out of it and leave us to resolve the crime problem through non-criminal-justice means. I think both perspectives miss out on an important dimension: it doesn’t just matter how much policing is taking place, but also what kind of policing.

William Muir’s 1977 classic Police: Streetcorner Politicians offers a matrix that characterizes police officers based on their psychology. Muir is interested in two dimensions: the officer’s proactivity and their worldview. These create four types of cops.

ProactiveReactive
TragicProfessionalReciprocator
CynicalEnforcerAvoider
Adapted from: William Muir, Police: Streetcorner Politicians (Chicago, 1977)

Out of these four, Muir’s preference is for the professional, whom he sees as an energetic, passionate problem solver who has compassion for their community. But preferring the professional to other types depends on the extent to which one believes that cops can still have a tragic/empathetic approach to human nature and the human condition. People who assume that all cops are cynical about the people they serve face a choice between enforcers and avoiders and might prefer avoiders. People who believe that some cops can be professional and compassionate, will prefer professionals to reciprocators.

If Muir’s typology is not applied to individual cops, but rather to hypothetical cops as “ideal types” of what we would and would not like to see in the streets, I think the best way to understand the Arab-Israeli call for help is as a call for professionals, not for enforcers. Which raises the question, given that we pay taxes so that we can have police services, why not insist that the force hire professionals rather than enforcers? Why give up and settle for avoiders, or for shrinking the force (and its utility) altogether? How much despair people experience and, consequently, how much they believe that they can have the police force they deserve, could be (as I argued in the previous installment) a function of where they live or (as I argue in this installment) on what moment we are in.

The Call Is Coming 2: A Comparative View of Approaches to Intra-Racial Crime

In the last few days, I’ve been thinking a lot about the issue I brought up in my previous post The Call Is Coming from Inside the House–situations in which a minority community with an appalling history of oppression by law enforcement asks for law enforcement intervention due to rising crime rates. I compared the Arab Israeli protests for law enforcement intervention to the letters recently penned by NAACP leaderships in Oakland and San Francisco.

What I didn’t discuss was an obvious difference in the way these calls for enhanced law enforcement reverberate within these communities and outside them, which puzzles me. As I explained in that post, the Arab Israeli community is fairly united in its call for police intervention and personal safety. The pressure on allies and other members of the center-left opposition to Netanyahu is to participate in protests calling for the police to investigate and solve these crimes. People get excoriated for not embracing this call.

By contrast, the NAACP calls I looked at in the previous post have by no means represented the consensus in the Black community and, in fact, provoked a lot of strife and antagonism. There is serious critique and questioning of the concept of “Black-on-Black crime”, efforts to present police violence as a much more important and salient problem than the crime problem, and pretty oppressive silencing of the few white voices that don’t fall in line with the “don’t call 911” ethos.

This difference in approaches is striking not only within minority groups and their adjacent communities, but also among academic and human rights milieus. In Israel, ACRI (the equivalent of the ACLU) feels deeply conflicted on what to do and who to support. And in the U.S., academics and nonprofits by and large fall in line with the idea that the priority is to curb police violence, rather than intra-racial civilian violence.

At the recommendation of a friend, I started listening to Micha Goodman and Efrat Shapira-Rosenberg’s podcast Miflegeth HaMahshavot (“The Party of Thoughts”), which explains ideologies in Israeli politics. In one early episode, they explain the rise of Ra’am, the first time a major Arab party was part of the Knesset. According to Goldman, this election represents the triumph of sectorial interests, which Ra’am sought to promote, over the big issue of the Palestinian occupation that the Joint Party, the other Arab party, sought to promote. Ironically, though, Ra’am is an Islamist party, presumably less inclined toward compromises, which raises the question how it came to offer Arab-Israeli voters a pragmatist, sectorial platform. Goodman thinks that it reflects a unique form of religious pragmatism: we, humans, worry about our immediate, short-term issues (chief among which is the intraracial crime problem), while God/Allah will worry about our ultimate salvation (an Arab state from Jordan to the Mediterranean sea).

If applied to the U.S., Goldman’s might predict a similar sectorial emphasis on restoring personal safety to the neighborhoods referred to in Supreme Court jurisprudence “high crime areas” and in sociological parlance “neighborhoods where poor people of color live.” And yet, that’s not what we’re seeing. Either fighting crime is not (or, until recently, was not) a sectorial issue of high priority for Black communities, or police violence is more of an issue of that sort. Why are we not seeing parallel processes in the two countries, then? hypothesis would So, why is there a difference?

I’ve tried to hash this with friends, and I’m not sure I’ve nailed the issue, though I have some thoughts. Let’s work through this the way Hercule Poirot would solve a crime: by gathering suspects and eliminating them from our inquiries. The first two possibilities are related to the with the relevant weight of the crime and police problems in the two countries, and I find both unpersuasive:

  1. The crime problem and the threat to personal safety are much more serious in Israel than in the U.S. This is not something that is easy to measure, and geography makes a big difference. Crime is not evenly distributed in either country. The existence of “million dollar blocks” and places ravaged by gang warfare is unfortunate, but not fictional. I think in both places there are people living under a serious threat of violent crime.
  2. Police violence toward minorities is a much more serious problem in the U.S. than it is in Israel. This is also something that is difficult to measure, especially due to problems of underreporting. Again, geography makes a big difference, because in both countries enforcement is selective and very geography-driven. In addition, the national security/conflict in Israel throws in another factor (there are now voices calling to involve Israel’s security service, the Shabak, in crime solving in Arab Israeli villages. Yikes.) I would have to parse out the statistics, but I don’t see that Arab Israelis are more fortunate than Black Americans in the treatment they receive from the police.

If we accept the premise that crime rates and police violence are serious problems for both populations in both countries, we should consider the extent to which the crime picture emerging from the two context is different. In other words, can Arab-Israeli crime be distinguished from Israeli crime in general to the degree that Black crime can be disaggregated from American crime? How easy is it to treat it as a unique, endemic problem? Again, two options emerge, one sociological and one involving framing.

  1. The sociological issue: Perhaps voices in the Arab-Israeli community are more successful in raising crime rates as a problem because intra-racial violent crime in Israel is, or is perceived as, more of a stereotypically/characteristically Arab/Palestinian problem than intra-racial violent crime in the U.S. is perceived as a stereotypically Black problem. This requires viewing murder cases, including unsolved murders, through a criminological lens. I have the 2021 data. What it tells us is that Arab- Israeli murders might not be as distinctive as the media suggests. In a previous post I described the disturbing statistics about the murder of Arab women, but those are less than 13% of overall murders in the Arab community. We know most of these are shoot-outs and most of the victims are under 30 years old. This doesn’t seem to paint a picture full of honor killings and, in fact, resembles organized crime killings in the U.S. Both countries also feature problems involving the proliferation of guns in criminal hands that are certainly not limited to this or that ethnic/racial group. It is true that, in Israel, 64% of murder cases are perpetrated by Arabs (usually against Arabs), who are merely 21.1% of the general population. FBI UCR data for 2019 shows that African-Americans (who were 14% of the U.S. population in 2021) accounted for 55.9% of all homicide offenders in 2019. In both cases we have considerable overrepresentation that cannot be explained merely by discriminatory policing/investigatory practices.
  2. The framing issue: Perhaps politicians on the left in Israel feel more comfortable calling for police intervention to solve intra-racial crime in Israel because there it is not perceived as being tied to, or stemming from, the Jewish/Zionist hegemony to the extent that intra-racial crime in the U.S. is perceived as a response to white supremacy. Even if this is true, it raises a further question: what impacts the framing?

Which brings us to the final frontier: I think that a big difference between Israel and the U.S. has to do with intra-movement politics and positionality, and these factors are responsible for how the problem is framed:

  1. I think that Goodman is right in that Israeli Palestinians/Arabs have become more invested in sectorial politics, while the U.S. Black community has by-and-large retained its interest in the bigger questions of criminalization/incarceration.
  2. This could be related to the respective size of the two countries in two ways. First, in Israel there’s more segregation in terms of where people live. This means that educated, middle-class Arab Israelis will live in closer proximity to crime than middle-class Black Americans and, because of that, will be more invested in personal safety and law enforcement (this is in line with James Forman’s argument about D.C., which is a city in which Black politicians and police officers hold considerable power and use it to “lock up their own.”) Second, the sheer population of the minority group is so much smaller in Israel that, to the extent that someone even cares about the plight of the community, it will hear mostly from middle-class, law-abiding folks afraid to let their kids outside to play; in the U.S. there’s a multiplicity of voices which, amplified by social media and activism, includes the interests of those more concerned about police persecution than about crime prevention.
  3. Finally, I think the Israeli scenario contains an important factor: Arab/Palestinians who are Israeli citizens are in a completely different situation than Palestinians living in Palestine. The latter are in such dire straits, and treated so appallingly by the army, the security services, and the settlers, that the police-citizen encounters against Israeli citizens, ugly as they may be, don’t even register as a problem by comparison.

The Prospect of Freedom

It looks like the event I hardly imagined could ever happen might happen: Gov. Newsom announced that he would not appeal the Court of Appeal’s decision to reinstate Leslie Van Houten’s parole. The Associated Press reported Newsom’s obligatory statement, as well as some valuable words from Van Houten’s attorney:

In a brief statement, the governor’s office said it was unlikely that the state’s high court would consider an appeal of a lower court ruling that Van Houten should be released.

Gov. Gavin Newsom is disappointed, the statement said.

“More than 50 years after the Manson cult committed these brutal killings, the victims’ families still feel the impact,” the statement said.

. . .

“She’s thrilled and she’s overwhelmed,” [her attorney Nancy] Tetreault said.

“She’s just grateful that people are recognizing that she’s not the same person that she was when she committed the murders,” she said.

After she’s released, Van Houten will spend about a year in a halfway house, learning basic life skills such as how to go to the grocery and get a debit card, Tetreault said.

“She’s been in prison for 53 years. … She just needs to learn how to use an ATM machine, let alone a cell phone, let alone a computer,” her attorney said.

In the last chapter of Yesterday’s Monsters I looked at factors that might or might not be conducive to the release of any of the people I featured in the book. What paved the way to Van Houten’s release was the reemergence of adolescence as a relevant factor for parole; she had been living an exemplary life of self-reflection and rehabilitation for many years and was nonetheless repeatedly denied based on “lack of insight”, which, as I and others have discussed, is nonfalsifiable. The Commissioners had always known that Van Houten was 19 when she participated in the LaBianca murders, but a series of Supreme Court decisions gave them explicit permission to consider her age, and that’s what tipped the scales.

I now see these cases through the broader prism that Chad Goerzen and I develop in FESTER (coming to a store near you in January 2024), which highlights the travesty of turning our prisons into geriatric facilities. Beyond the obvious issue of medical risk, there is the question of what freedom looks like to a septuagenarian leaving prison after 53 years. Even without the added difficulty of immense stigma and animosity, Van Houten, who is a bright, thoughtful, and talented person, and who has academic degrees, will find it very hard to find and hold a job in this market. She has had to endure a lot behind bars, and her reentry is unlikely to be very easy. I wish her all the best. Our paths did not cross–she did not wish to be interviewed for Yesterdays’ Monsters–but I very much hope they will some day.

The Horrid, and 100% Foreseeable, Aftermath of Judge Persky’s Recall

It is the nature of character assassination scandals, and a consequence of their frequency, that after a while they are forgotten by all except the people whose lives were destroyed by them. Such was the fate of Michelle Dauber’s cruel and idiotic crusade against Judge Aaron Persky in the aftermath of the Brock Turner scandal, which swept a lot of ill-informed progressive punitives with pitchforks and led to the destruction of his judicial career (and later, the destruction of his livelihood as a tennis coach.)

For all the shrill shrieking about “privilege”, pretty much every criminal justice academic I respect in the Bay Area warned at the time that recalling judges for lenient sentencing (especially, as in this case, following the probation recommendation) would make punishment harsher and much worse for everyone–especially for people who looked and lived nothing like Brock Turner. I was one of the first signatories and vividly remember shouting this from the rooftops, as well as seeing it as part of an appalling pattern of the left eating its own with no rhyme or reason.

As everyone worth their salt predicted, the recall did have an effect on criminal punishment in Santa Clara county: it made it harsher. As my colleagues Sanford Gordon and Sidak Yntiso found:

Using disposition data from six California counties and arrest records for a subset of defendants, we find a large, discontinuous increase in sentencing severity associated with the recall campaign’s announcement. Additional tests suggest that the observed shift may be attributed to changes in judicial preferences over sentencing and not strategic adjustment by prosecutors. We also demonstrate that the heterogeneous effects of the announcement did not mitigate preexisting racial disparities. Our findings are the first to document the incentive effects of recall and suggest that targeted political campaigns may have far-reaching, unintended consequences.

Sanford C. Gordon and Sidak Yntiso, Incentive Effects of Recall Elections: Evidence from Criminal Sentencing in California Courts, The Journal of Politics 84:4 (2022), 1947-1962

In other words: the fearless, plucky lefties who led this hysterical campaign can take pride in the fact that their relentless persecution of Persky empowered and enhanced carceral repression across the board, not necessarily making a dent in prevention/accountability for sex crimes, and harming precisely the people without “privilege” that they presumably sought to protect with this destructive campaign.

Why am I revisiting this? For two reasons. First, because I don’t want us to forget that these sorts of actions have consequences. I know that many on the left are already sickened by years of ugly, disastrous infighting. As Freddie DeBoer recently wrote:

I certainly would not say that the age of canceling is over. There will be public scandals to come; people will suffer major career and social consequences because of public anger. Sometimes they’ll deserve it. And maybe this is just a lull and the same old songs will get sung again and again.

But at this stage I find it hard to deny that the sense of palpable fear so many operated under, the feeling that the prosecutors held all the cards, appears to be in terminal decline. People just aren’t afraid in the same way anymore. The mob doesn’t have the momentum. The big bad wolf has lost his teeth. I suspect this is for a few key reasons – the fact that all of that endless raging did precisely nothing to make the world more just, for one. The growing understanding that the human species is flawed by nature and that no one can match those standards, for another. But mostly, I think it’s the dynamic I’ve been predicting for a long time: you can only bang the gong so many times. Everybody’s receptors all got blown out. Outrage is a finite resource. People can’t maintain permanent offense forever. Most of us can’t, anyway. You can only tense a muscle for so long.

Thing is, while we are tiring of the phenomenon, the people whose lives and reputations were laid to waste are going to have to live with the consequences of these witch hunts for a long time. Judge Persky, I think of you and am so sorry for the horror that you went through.

The second reason is that an excellent, short documentary about the poisonous effect of the recall campaign is out, and until September you can watch The Recall: Reframed for free:

Op-Ed in the L.A. Times Critiquing the Gubernatorial Veto on Parole

Today, the L.A. Times published my op-ed, in which I criticize California’s gubernatorial veto on parole which, as I explain in Yesterday’s Monsters, serves no purpose except contaminating the parole process with politics and optics. Here it is:

***

On Tuesday, California’s 2nd District Court of Appeal reversed Gov. Gavin Newsom’s veto of Leslie Van Houten’s parole, reinstating the state board’s parole grant decision. Their ruling exposes deep flaws in California’s system of allowing gubernatorial vetoes in the first place.

Van Houten, a member of the infamous Manson “family,” participated in the murders of Rosemary and Leno LaBianca in 1969. She was 19 at the time. These were horrific crimes whose aftermath shattered a sense of innocence and safety for many. But it is also true that Van Houten and other teenage girls caught in Manson’s web were indoctrinated into, exploited and abused by a dangerous cult not properly understood until many years after the murders.

In prison since 1971, with her original death sentence commuted to life with parole in 1972, Van Houten has transformed herself, earning two academic degrees, participating in rehabilitative programs and expressing remorse for her crimes. After decades of prosecutors and families of the victims of Manson’s crimes opposing Van Houten’s release, the factual evidence finally outdid the political pressure: Since 2016, the Board of Parole Hearings has recommended her release five times. Gov. Jerry Brown and then Gov. Newsom reversed each decision.

The appeals court reviewed the veto through a system deferential to the governor; all they needed to uphold his decision was “some evidence” that Van Houten, now 73, presents a risk to public safety. The court concluded that his veto was “not supported by a modicum of evidence in the record.”

Since a 2008 decision from the California Supreme Court, parole boards can’t deny release based solely on the severity of a crime. Instead, they must show that the parole candidate poses a public safety risk. Boards and governors alike have circumvented this standard by using hard-to-falsify language — for example, vaguely claiming that they don’t think the inmate possesses “insight” about their crime.

In denying Van Houten’s 2020 parole bid, as the appeals court reported, Gov. Newsom argued that her “explanation of what allowed her to be vulnerable to Mr. Manson’s influence remains unsatisfying.” He was also “unconvinced” that her childhood trauma, including her parents’ divorce and a forced abortion, “adequately explain her eagerness to submit to a dangerous cult leader or her desire to please Mr. Manson, including engaging in the brutal actions of the life crime.”

The court essentially called the governor’s bluff. They found that Van Houten’s extensive record showed “no additional factors Van Houten has failed to articulate, or what further evidence she could have provided to establish her suitability for parole. The Governor’s concern that there is more than meets the eye is, on this record, speculation, but [per state law] the Governor’s ‘decisions must be supported by some evidence, not merely by a hunch or intuition.’”

Yet allowing the governor to veto parole recommendations at all risks reducing such weighty decisions to one person’s hunch or political agenda. California is one of only two states that allow gubernatorial veto of parole. The Legislature introduced it in 1988, politicizing the parole process and adding public pressure — as well as optics — to what should be a professional assessment of risk. The veto works in one direction: The governor can only veto parole recommendations, not denials.

Any fear that the state is releasing dangerous people in droves is unfounded. Parole boards are reluctant to grant parole. According to data from the California Department of Corrections and Rehabilitation, the Board of Parole Hearings recommended it in only 20% of cases in 2019. As I explain in my book “Yesterday’s Monsters,” receiving parole at one’s first hearing is extremely rare. I found that the median time spent behind bars on a life sentence with parole in California has risen from 12 years in 1980 to 28 years in 2012 for those who have been released, and a quarter of the prison population is serving life sentences — 26,000 with parole and 5,000 without.

The role of politics was particularly clear during the COVID-19 pandemic. The aging and infirm lifer population faced serious risks of contagion and death behind bars. They also pose little to no public safety risk, as shown by robust criminological evidence. Still, Newsom agreed to release merely 8,000 people — a deficit eclipsed by incoming admissions from jails, and the vast majority with just weeks or months left of their sentences. Van Houten was up for parole in 2020 when her prison, the California Institution for Women, was experiencing a COVID-19 outbreak of more than 100 cases.

The court’s decision now puts the ball back in the governor’s court. He has a 10-day window, starting in a month, wherein he can instruct Atty. Gen. Rob Bonta to appeal this case to the California Supreme Court. Common sense should prevail and guide our leadership in Sacramento to allow this rehabilitated septuagenarian to live her life quietly on the outside.

But no matter the outcome, her journey raises serious questions about the gubernatorial veto. Do we truly need an extra layer of political considerations to assess danger to the public — or should we trust the professionals appointed by the governor, mostly from law enforcement backgrounds, to do their job?

Hadar Aviram is a professor at UC Law San Francisco. She is the author of “Yesterday’s Monsters: The Manson Family Cases and the Illusion of Parole” and co-author with Chad Goerzen of the forthcoming “FESTER: Carceral Permeability and California’s COVID-19 Correctional Disaster.”

CA Court of Appeal Reverses Newsom’s Veto, Reinstates Van Houten’s Parole

Parole drama today! the California Court of Appeal for the Second District reversed Gavin Newsom’s veto of Leslie Van Houten’s parole, reinstating the board’s parole grant decision.

You can read the decision in full at this link. It’s a 2:1 decision, with the majority opinion recounting Van Houten’s early history, life crime, prison history, and plans for release. Their point of departure is a standard of review that is highly deferential to the governor:

We review the Governor’s decision under the “some evidence” standard, a standard our Supreme Court has called “extremely deferential.” (In re Rosenkrantz (2002) 29 Cal.4th 616, 665 (Rosenkrantz).) Under that standard, a simple modicum of evidence is all that is required to uphold the Governor’s decision. (Shaputis, supra, 53 Cal.4th at p. 210.) “Only when the evidence reflecting the inmate’s present risk to public safety leads to but one conclusion may a court overturn a contrary decision by . . . the Governor.” (Id. at p. 211.)

p. 43

But even under this standard of review, “we nonetheless conclude that the Governor’s reversal in this case is not supported by a modicum of evidence in the record.” (p. 44).

Their support for this assertion echoes what I said in chapter 4 of Yesterday’s Monsters: that the constant refrain that Van Houten has somehow failed to do even deeper psychic excavation into her circumstances and crime is nothing but smoke and mirrors, that it is ridiculous especially in someone so introspective, and that it is thin cover for political optics.

The Governor found that Van Houten’s “explanation of what allowed her to be vulnerable to Mr. Manson’s influence remains unsatisfying,” and he was “unconvinced” that Van Houten’s parents’ divorce and her forced abortion “adequately explain her eagerness to submit to a dangerous cult leader or her desire to please Mr. Manson, including engaging in the brutal actions of the life crime.”

p. 44

To which I’d say, how exactly does he expect anyone to explain a bizarre stranger homicide in the context of a cult? Is there anything she could possibly say that would lead anyone in Sacramento to write, “aha, now I get it”? The Court agreed, walking us through Van Houten’s introspection in a way that shows the Governor’s reasoning for the sham that it is:

It cannot be said that Van Houten has not extensively identified and discussed the factors leading to her life crimes, only some of which briefly are referenced in the Governor’s decision. In both her interview with the CRA evaluator and at the parole hearing, Van Houten expounded at length on the causative factors, beginning with her feelings of anger and abandonment after her parents’ divorce, a stigmatizing event in that era, and how that led to drug and alcohol abuse. She ran away from home with her boyfriend, who had impregnated her. Her mother then forced her to have an illegal abortion against her wishes, unmedicated, in her bedroom, instructed to keep quiet so as to not wake her siblings.

Van Houten spoke of shutting down emotionally and feeling numb after the abortion. The CRA evaluator wrote that, even now, Van Houten “was tearful as she spoke of the abortion and what ‘might have been.’ ” Van Houten described herself at that point in time as being “ ‘[d]esperate to be accepted,’ ” and “ ‘ha[ving] no sense of value. My value came in the eyes of other people.’ ”

Van Houten stated when she met Manson cult member Catherine Share, she “was at an all-time bottom low. I had no income, I did not feel good about either of my parents, and when I met her, it seemed to me that I was being offered a pretty good life.” She described how Manson slowly indoctrinated her, often while she was under the influence of LSD. The cult was not murderous and violent at the outset—rather, she stated her time at the ranch initially “ ‘seemed fun,’ ” and the talk of and preparation for violence and revolution came later. Van Houten said she “ ‘wanted to belong and . . . wanted to belong to something that wasn’t connected to my past.’ ” Van Houten explained how Manson used her anger with her parents and her shame about the abortion to convince her to turn her back on society, accept the alternative lifestyle he offered, and reject the lessons of right and wrong she had learned in her youth. Manson successfully transformed any doubts Van Houten had about the cult into her own self-criticism for failing to achieve the enlightenment he purportedly offered. By the time Manson’s talk turned to violence and murder, Van Houten already had fully committed to him, so much so that she believed he was Christ reborn. She also believed in the impending revolution, and that remaining with Manson was key to her survival.

The Governor found Van Houten’s extensive discussion of the causative factors inadequate to explain her life crimes. This necessarily implies the Governor believes there are additional factors for which Van Houten has failed to account, factors that, unaddressed, create a risk of violent recidivism. There is no indication in the record, however, of a latent underlying factor that potentially could result in violent conduct, nor has the Governor identified one. The CRA evaluator found Van Houten did not meet the criteria for psychopathy or a personality disorder, and there was no evidence of a thought disorder, hallucinations, or homicidal or suicidal thoughts or behavior. The evaluator further found it “very likely” that Van Houten’s youth at the time “significantly impacted” her involvement in the life offense, a factor obviously no longer applicable five decades later. The CRA’s finding that Van Houten presented a low risk of recidivism was consistent with similar evaluations over many years. Van Houten, moreover, has no history of violence either before the life crimes or in the 50 years since, and the prison staff regarded her highly enough to place her in positions of leadership within the prison, including facilitating groups intended to help other inmates with their rehabilitation.

The record shows no additional factors Van Houten has failed to articulate, or what further evidence she could have provided to establish her suitability for parole. The Governor’s concern that there is more than meets the eye is, on this record, speculation, but the Governor’s “decisions must be supported by some evidence, not merely by a hunch or intuition.” (Lawrence, supra, 44 Cal.4th at p. 1213.)

pp. 45-47

The unwritten part of this is pretty obvious to me: the only factor that can explain this veto is political optics, and California law does not allow Governors to veto people’s parole because it will look bad and people will write mean things on Twitter.

The only remaining question is: What happens next? It is quite possible that Gov. Newsom will instruct Attorney General Bonta to appeal this to the California Supreme Court and to ask for an en banc decision, which will further delay proceedings. I was asked today whether they could hold Van Houten in prison while they do that. I honestly am not sure. I will say, though, that if she’s released pending the CA Supreme Court decision, it’ll be the second time she’s spent some time on the outside–this is what happened when her request for a new trial was approved in 1979.

There is another reason why Newsom’s decision was outrageous: you may not remember this, but when Van Houten’s case was pending before Newsom, CIW, where she is incarcerated, had a horrendous COVID-19 outbreak. As I wrote at the time, to keep a 72-year-old woman in prison when she has no disciplinary record whatsoever and is lauded and appreciated for her superb behavior and personal growth at a time when her congregated facility has a huge outbreak was inhumane. I really hope our leaders in Sacramento can let go of ego and optics, set aside their personal aspirations and dread of negative publicity, and do what is unquestionably the right thing here: let this go and allow this low-risk septuagenarian woman with advanced degrees to live her quiet life on the outside.

Cooperation and Disruption in True-Crime Podcasting: Your Own Backyard

As some of you know, I’m beginning to work on a new project that sits at the intersection of new media, victimology, and law enforcement. I’m interested in the true crime podcast community, especially in podcasts targeting unsolved crimes. One of the questions I’m deeply interested in is the give-and-take between official law enforcement and podcasters (whether family members of the victim or third parties), which seems to range from hostility, through begrudging acceptance, all the way to pretty warm cooperation.

One notable example is Chris Lambert‘s excellent podcast Your Own Backyard, which is a thorough investigation of the disappearance of Cal Poly student Kristin Smart in May 1996. Lambert, who started off as an absolute stranger but established a warm collaborative relationship with the Smart family, has produced a true investigative masterpiece, chock-full of resources, first-hand testimony, circumstantial evidence examination, and intelligent inquiry into various forensic science disciplines (including human remains dogs and ground-penetrating radar). Most remarkably, Lambert’s podcast not only reawakened public interest in Smart’s disappearance, but also brought in new witnesses from the woodwork. Lambert’s dogged perseverance, intelligent analysis, and commitment to finding out the truth earned him the trust of the surrounding community and of law enforcement, and it looks like the police greatly benefitted from his work.

Throughout the entire lifespan of the case, there was only one viable suspect in Smart’s disappearance: fellow student Paul Flores, who helped an inebriated Smart get home from a party and was the last person to see her alive. Flores and his parents acted evasively and suspiciously over the years; Lambert’s investigation revealed that Flores was a predator who made women uncomfortable before Smart’s disappearance and, years after the event, a prolific rapist of multiple women. As Lambert provocatively posited in the podcast, Flores would have to be the unluckiest man alive for Smart’s disappearance to have been a coincidence.

Smart’s body was never found, but there was some evidence of human remains at Flores’ father’s house. The San Luis Obispo DA decided (thanks in great part to Lambert’s work and the evidence unearthed by the podcast) to charge Flores with murder and his father with being an accessory after the fact (to solve the confrontation problems in trials with codefendants, there were two different juries attending the same trial; I can talk more about this method, and how effective it is in solving Bruton/Gray/Cruz confrontation problems, in a future post). In March, the jury convicted Flores of the first-degree murder of Smart, and he was sentenced to 25-years-to-life in prison. This is a remarkable result given the passage of time and the hurdles in prosecuting no-body homicides.

I recommend listening to the whole podcast–it’s truly one of the better exemplars of this genre. One of the many things I find interesting, though, is the extent to which the existence of the podcast and its centrality to the case played a part in the criminal trial. In an effort to remain objective, Lambert, who recounts the trial in the later podcast episodes, matter-of-factly reports courtroom mentions of his own podcast without editorializing. But the defense (as a defense attorney, I gotta give kudos to Robert Sanger for what I think is undoubtedly a pretty heroic showing of professionalism with a client who is a pure, unadulterated garbage of a human being) repeatedly refers to the podcast and its encroachment on the case. Witnesses are asked about their participation in performative support for the Smarts (such as the entire investigative and prosecutorial team wearing purple, Smart’s favorite color) and about the extent to which the podcast propelled them to step forward. I’m pretty sure there will be arguments aplenty about bias and prejudice on appeal, and I worry that the podcast’s huge contribution to the investigation will seriously backfire.

Which brings me to one of my concerns about new media and law enforcement in general: Overall, I’ve been really impressed with the power of podcasts, especially their contribution to diversifying and enriching the victims’ rights movement. But is it time to have a sit-down, perhaps at CrimeCon, and set up some ethical rules, or best practices? Not everyone is Sarah Turney or Chris Lambert, not everyone does their homework in a dogged, meticulous way, and I worry that the need to come up with provocative encounters, confront suspects, dig up drama, etc., might backfire especially when podcasts finally succeed in greasing the wheels of the criminal process. Some things I think are worth considering are:

At what point should podcasters who are not themselves related to the victim reach out to the victim’s family? Is it ever okay to produce a podcast that the victim’s family does not support? What if the podcast casts suspicion on the family itself?

What kind of relationship should podcasters foster with the police? At what point should they hand evidence over to the police? Is this relationship akin to the police’s communication with traditional journalists?

Who owns footage obtained and produced by podcasters? Is there ever some sort of evidentiary privilege akin to the one granted to traditional journalists?

How much verification is required from podcasters (say, by contrast to police detectives checking alibis or triangulating evidence)?

What are the rules of engagement when reaching out to suspects? If podcasters take risks, how, and to what extent, does the police need to support and protect them–especially when law enforcement does not think that confronting the suspects is prudent?

Do podcasters have responsibility for the public chatter generated around the podcast? Wild theories, blame casting, and garden-variety shitposting that might happen, including, for example, posts that disparage the victim and/or their family?

What are the considerations that govern the way in which the story is told? For example, is it ethical to refrain from disclosing certain incidents/developments out of artistic concerns, or to make the narrative more dramatic and engaging? And what about the tone of reportage? Some of these podcasts (emphatically, NOT Lambert’s or Turney’s) have a humorous, flippant tone–is that something that should be frowned upon, especially if the victims’ families are not on board?

I’m interested to hear from you what other concerns/thoughts you have about these podcasts. And let’s keep tabs on the appellate process in the Flores case.


Comment: I’m still in Israel by my dad’s bedside – I write just to have a placeholder for ideas that pop in my head during my morning run before I head to the hospital every day. Please, no cumbersome professional requests during this trying time for me and my family.

An All-Male Jury for a Groper and the G2i Problem

The Gemara relates: Rav bar Sherevya had a trial pending before Rav Pappa. Rav Pappa seated him and also seated his litigant counterpart, who was an am ha’aretz (a simple man, not a rabbi). An agent of the court came and kicked and stood the am ha’aretz on his feet to show deference to the Torah scholars there, and Rav Pappa did not say to him: Sit. The Gemara asks: How did Rav Pappa act in that manner by not instructing the am ha’aretz to sit again? But aren’t the claims of the am ha’aretz suppressed by Rav Pappa’s perceived preferential treatment of Rav bar Sherevya? The Gemara responds: Rav Pappa said to himself that the litigant will not perceive bias, as he says: The judge seated me; it is the agent of the court who is displeased with me and compelled me to stand.

Shevuot 30b

Understandable outrage is brewing among many folks around me: At a San Francisco trial of a man accused of stalking and groping women, all the jurors are male. How could this happen? And is it lawful? Let’s go over some terminology:

  • Population: everyone who lives in the county.
  • Sampling frame: the group of people from which one can draw a sample. For our purposes, the folks whom the law deems eligible to serve on juries in the county.
  • Venire: Everyone who received summons to appear for jury selection (the selection process itself is called “voir dire.”)
  • Panel: The people who are eventually seated on a particular jury.

The constitution requires that the jury be drawn from a “fair cross-section” of the population: in other words, that the jury pool–the overall sampling frame from which people are summoned for the venire–be reflective of the population. If some recognizable minority group is systematically disqualified from serving, the selection method is unconstitutional. In the landmark case Taylor v. Louisiana, the Supreme Court invalidated a jury selection scheme by which women were not summoned at all to the jury pool unless they explicitly chose to opt in. Similarly, schemes like Texas’ “key man” system, where there’s some official who gets to pick and choose who’s on the jury (and thus, for example, underrepresents Mexican-Americans) have been invalidated.

Having a sufficiently diverse jury pool, however, does not guarantee the empaneling of a diverse jury. Consider the following example: you have 100 pebbles, 50 of which are gray and 50 of which are purple.

The statistical odds of drawing a purple pebble are 0.5, which means that, in a random selection of 12 pebbles, the stats predict you have great odds of having a mix of gray and purple pebbles. But you can easily imagine many random drawings that will only include gray pebbles.

This is exactly what happened here, except for an important fact: the twelve jury members were not drawn at random. Annie Vainshtein and Nora Mishanec reported for the Chron:

During jury selection, some women said they could not impartially weigh the evidence that would be presented at trial due to personal experiences with sexual assault or harassment, or negative feelings toward Hobbs’ attorney, which prompted Superior Court Judge Harry Dorfman to dismiss them.

Others from the pool were unable to serve on the jury for different reasons; one woman said she had booked an upcoming cruise. Several jurors, one of whom was male, were dismissed after expressing opinions including that “sexual predators” should be segregated from society, and even face the death penalty. 

By the end of jury selection, the only woman selected was an alternate juror, who will hear all of the evidence but vote on the verdict only if needed.

Here’s the thing: robust social science research tells us that, when looking at groups of people in the aggregate, people’s life experiences and worldviews, which are often a function of their demographics, impact how they will assess evidence and judge a case. Which is why, even without resorting to the services of expensive trial consultants, prosecutors assume that people of color will be favorable to the defense, and defense attorneys assume that white men will be more punitive. The name of the game in voir dire then becomes getting rid of as many people whom you suspect will be unfavorable to your side. The problem is that, even though we can make these generalizations regarding groups, we have a deep social distaste about making them regarding individuals: people generally recoil from being told that they must think a certain way because of who they are, even though in the aggregate we know such statements to be true. This is why one can’t mount a for-cause challenge for disqualifying a woman, any woman, from the trial of an alleged stalker/groper just on the basis of her sex/gender. In science, it’s known as the group-to-individual (G2i) problem, and it affects various areas of legal decisionmaking.

Over the years, parties have tried to skirt this problem by using peremptory challenges to get rid of demographics they suspected of being unfavorable to them; the advantage of this strategy was that peremptories didn’t require an explanation. But the Batson doctrine allows the opposite party to challenge such use of peremptory challenges when they reveal a pattern of discrimination against a suspect racial or gendered group. It used to be the case that all the prosecution had to do was provide a race neutral explanation for their challenges (which, admittedly, would be difficult if there was evidence to refute this.) Now, California’s new peremptory challenge laws, enacted through AB 3070, make it a lot more difficult to get away with this sort of thing, because the prosecution’s explanation has to be reasonable, and it also cannot correlate with a seemingly race-neutral explanation that strongly correlates with race, gender, or any other suspect category.

But this is not what happened here! The women were dismissed using for-cause challenges because they directly opined that they would not be able to impartially weigh the evidence. This I find dubious (though not impossible) and it leaves me with serious discomfort. To drive home the problem, consider the following analogy: assume a white police officer is on trial for shooting and killing an unarmed black man. Imagine that, at jury selection, every single black prospective juror says that they would not be able to impartially weigh the evidence and, consequently, we end up with an all-white jury. Does this pass the “fair cross section” test? Yes–there were people of various races in the jury pool. Does this pass the Batson test? Sure! No peremptory challenges were used; everyone who was struck was struck for cause. Are you comfortable with the outcome?

How could this have been fixed? First, I think that prospective jurors can and should trust their ability to make good decisions with the life experience that they have. Like 50% of the people on the planet, I have been sexually harassed, catcalled, groped, pestered for sex, and other fine experiences. Does that mean I would not be able to seriously consider the possibility that a person who did this to others was severely mentally ill, or that there was an eyewitness identification problem? I worry that the emphasis we put on group identity in contemporary discourse has locked people into beliefs that they are immutable members of whatever demographic they belong to and there’s nothing more to them, and that is impoverishing and disappointing. Second, I think the onus here was on the prosecution to ask the prospective jurors questions that would probe the extent of the bias. For example, I think a fair question would have been, “would your experience with harassment lead you to find someone guilty even if there was defense evidence that the police got the wrong person, or even if there was persuasive psychiatric evidence that the defendant didn’t know what he was doing?”

If such a stunning number of women find themselves unable to fairly adjudicate a sexual harassment case, then the root of the problem here is not the jury selection process itself. It is the fact that harassment experiences in public space are so common and far more malignant than people think. In her book License to Harass, my colleague Laura Beth Nielsen exposes the unbearable lightness of offensive speech in public space and the many insidious ways in which it affects people’s everyday lives and decisions. It turns out that even behaviors that might not be a big deal on a one-off basis can add up to the point that people are so fed up with them that they don’t feel they can be objective on a jury.

If that’s what happened here, it’s a damn shame. Because the irony is that the very fact that there are many other people like this guy (who maybe just yell obscenities, rather than grope, and thus completely escape public censure) is what makes it impossible to adjudicate this guy by a true jury of his peers, which should include women.

How to Address Pseudo-Police Lethal Force?

There’s understandable community upheaval about a recent tragedy that rocked downtown San Francisco: A security guard at a downtown Walgreens store shot and killed Banko Brown and, the D.A. decided, will not be facing criminal charges for homicide. In this CBS-5 story, I explain what is happening to the extent I can, not having seen the evidence.

Why is the D.A. not pressing charges? The D.A.’s office has issued a statement according to which, having viewed the store’s video footage of the incident, they find that “[t]he evidence clearly shows that the suspect believed he was in mortal danger and acted in self-defense” and that, while Brown’s killing was a “tragedy,” “[they] cannot bring forward charges when there is credible evidence of reasonable self-defense. Doing so would be unethical and create false hope for a successful prosecution.”

How do they establish if someone acted in reasonable self defense? According to California’s model jury instructions (CALCRIM), a defendant prevails on self defense if they used force against another person while (1) reasonably believing that they, or someone else, “was in imminent danger of suffering bodily injury,” (2) reasonably believing “that the immediate use of force was necessary to defend against that danger,” and (3) “used no more force than was reasonably necessary to defend against that danger.” Note that it’s not enough that someone subjectively believes they are in danger; you prevail on this defense if a reasonable person in your shoes would’ve felt the same. The question of proportional force is also one on which there could be disagreement.

Hold on, why is this even a self defense issue and not a lethal force issue? The rules on lethal force apply only to government agents: federal and local police. The Fourth Amendment offers people protection from unreasonable search and seizure by the government. This is not the case here. The Walgreens security guard was, indeed, armed and working, but he was working for a private company. That’s why the D.A.’s office is viewing this incident through the lens of self defense, which applies to any altercation between two private people.

Shouldn’t we hold security guards and other pseudo-police officers, like private patrols, to a higher standard? I think that’s an excellent question, especially with the proliferation of private policing of all sorts. This is also far from the first time that someone was shot to death by private security personnel (see here, here, here, and here, to name just a few.) My friend Sarah Fielding, now the managing attorney at Legal Services of Northern California, once wrote a fantastic paper about neighborhoods who crowdfund for private security, and is interviewed in this fantastic Al Jazeera piece, where she and others express concerns that more exclusive, wealthy neighborhoods essentially just “send in a check” to fund their own justice, which disproportionally targets outsiders. All of these are serious problems that raise grave concerns. But the Fourth Amendment only applies to government agents and there’s precious little we can do about that.

If this is truly nothing more than a conflict between two private people, why all the political upheaval? Obviously, despite the fact that the shooter was not a police officer, the nature of this incident makes people see it through the lens of community outrage about police overreach. I’m seeing echoes of this even in Aaron Peskin’s reported entreat that the D.A. reconsider. People are falling into the predictable camps: pro-law-and-order, dismayed-of-downtown-chaos folks who support the D.A. and explain how sick they are of the robberies and petty thefts that have scared away many major retailers from the downtown areas, support the decision not to prosecute and see it as a victory for public safety. On the other hand, abolish-dismantle-repeal folks who, in accordance with the usual progressive punitivism protocol, think that we should not harshly enforce the law except against those we dislike (cops, guards, right wingers), see this as further evidence that Jenkins is making charging decisions that further oppress the oppressed. We don’t have the evidence, and so we are projecting our overall worldviews about the underlying problems of poverty, suffering, law enforcement, and dilapidation, onto this incident. This is where people’s strong views come from. The lesser the evidence, the more room there is for our worldviews to inform our imaginations.

We know there was only one gun at the scene–the guard’s. In light of this, isn’t it obvious that the security guard committed unjustified homicide? Folks, I don’t know. And neither do you, because we haven’t seen the video footage that the D.A. used to make the determination. For all we know, it might’ve seemed to the guard as if the victim was armed. Or not. We simply can’t know the answer to this without seeing the video.

Fine, then why won’t the D.A. show us the video, so we can be the judge of that? That’s a fair question. The understandable logic is: if it’s really that cut and dry that the security guard was in reasonable fear of his life, why is the D.A.’s office being so secretive about it? Because the public has deep interest in these issues, and in seeing justice done, California law was recently amended to require prosecutorial offices to share footage of lethal police shootings with the public within 45 days. But again, this doesn’t mean a general requirement to share investigative material with the public in any case that involves two private people, as the case is here. And, to be fair to the D.A., I can see some good reasons not to share the footage. We know from prior cases involving video footage of violence, sometimes lethal, that even when people have access to the evidence, their interpretations of what they see depend on their worldview. Fourteen years ago, I saw the cellphone footage of the killing of Oscar Grant and thought to myself, “I’m watching an execution, and there’s no way anyone could watch this video and think otherwise.” And, lo and behold, twelve Los Angeles residents disagreed with me. The D.A. might not want to get into these kinds of controversies if the law doesn’t require them to, nor do they want to incite confrontations and violence against the security guard. Still, it’s hard not to walk away from this with a bad feeling about the lack of transparency.

Then why not prosecute, and let a jury decide if it was self defense or not? Without seeing the video, it’s impossible to answer this question, but it’s definitely a fair one. I will say that the burden of bringing a case to trial is much lower than the burden required for conviction: all the prosecution needs is probable cause, as opposed to the much higher burden of beyond reasonable doubt. The idea is that you go to trial with a minimum of probable cause (the threshold the judge requires at a preliminary hearing) and continue developing the evidence as you go along. But in cases that are very public and sensitive, prosecutors sometimes hold themselves to a higher standard (e.g. Muller’s prosecutions of Manafort et al.) We should also keep in mind that prosecutors have two kinds of considerations that go into charging decisions: instrumental considerations–i.e., will this case end in a conviction and thus be a worthy expenditure of state effort and resources–and expressive considerations, i.e., what do my constituents want and expect. I think Jenkins is as aware of what her constituents want as Boudin was aware of his, and charging decisions will differ accordingly.

How Machine Learning Improves Parole Research

Rabbi Levi son of Rabbi says…The Holy One said to Moshe “You will make a menorah of pure gold” (Shemot 25:31).

Moshe responded: how will we make it?

God responded: “It will be made of hammered work” (Shemot 25:31).

But Moshe struggled and went down and forgot how to make it.

He went up again and said: My Master, how do we make it? God said: “It will be made of hammered work” (Shemot 25:31).

But Moshe struggled and went down and forgot.

He went back up and said: My Master, I forgot it!

God showed Moshe, and Moshe still struggled. God said to him: “See and create” (Shemot 25:40), and took a menorah of fire and showed him how it was made.

But, it was still a struggle for Moshe!

The Holy One said to Moshe: Go to Betzalel, and he will make it.

Moshe told Betzalel, and he immediately made it. Moshe was amazed and said: How many times did the Holy One show me, and I still struggled to make it! But you, who never saw it, knew how to make it by yourself!

BaMidbar Rabah 15

One of the professional events I most look forward to each spring is the Virtual Workshop on Contemporary Parole–a fantastic two-day online gathering of a rigorous group of people producing exceptional work, which we’ve now held for the third year in a row. The papers are always superb and so is the camaraderie and commentary. I got to present a draft version of my new Sirhan Sirhan paper, as well as hear really terrific work on various aspects of parole: gang validation, racial proxies, young adulthood, and others. I can’t go into too much detail, because these are all works in progress and we’ll probably see polished versions of everything getting published soon enough. But one thing that stood out to me was the uptick in really interesting work utilizing machine learning.

I know next to nothing about machine learning and, like Moshe in the midrash above, I might be too old a dog to learn that particular trick. I mean, in the Sirhan paper, n=1. Thing is, the midrash really resonates with me because I, too, feel a lot like Moshe when I hear someone else talk about a fantastic skill they have and how they put it to good use. It looks like, despite God’s repeated tutorials, Moshe’s goldsmithing skills weren’t up to snuff. Thankfully, there were other Israelites with that particular skillset: Betzalel was a gifted goldsmith who made a spectacular menorah on the first try (this is why Israel’s fantastic art school is named after him.) While unable to emulate Betzalel’s feat, Moshe had acquired a basic understanding of the necessary artistry and workmanship, so he could appreciate why Betzalel’s finished product was of such high quality. In other words–I don’t employ machine learning in my own work, but I know enough about it to be amazed when I read a paper that uses it well.

To understand the promise of machine learning, let’s first talk about how we do parole research the old-skool way. A multivariate regression works much like the denouement in an Agatha Christie mystery novel. You know the drill: Poirot gathers all the usual suspects in a room and goes through a litany of their motivations, opportunities, debunked alibis, you name it. He eliminates them one by one until he can point to the culprits. The important point is that Poirot selects who goes into the parlor for that last scene: people get there by invitation, and Christie is careful to craft the scene so that it’s pretty much always a finite and manageable list of people. When I run a regression, I pretty much do the same: I think about the dependent variable–the phenomenon I’m trying to explain–and I try to come up with a list of the independent variables that might explain it. For example, if my determinate variable is a parole grant, I ask myself: Do people who are represented by a private attorney do better than people who are represented by a panel attorney? Do people whose hearings happen in the morning fare better than folks who are heard in the afternoon? If victims and/or prosecutors show up for the hearing, does that make a difference? Does the professional background of the commissioners matter? Do people in some prisons stand a better chance of being granted parole? You can tell that each of these assumptions has a certain logic behind it (you get what you pay for; people are more attentive and in a better mood when they are not tired or hungry; professional background goes into constructing people’s worldviews; some prisons have better rehabilitative offerings than others, which improves one’s case.) I put all of these “suspects” in a room (the regression equation,) run the numbers, and see which comes out significant.

One of the problems with this model is that regression models rarely offer a complete and exhaustive prediction of the phenomenon they try to predict. There is even a statistic, the r-square, that measures how much of the dependent variable is explained by the set of independent variables we coded for. But there could be many factors that play into a parole grant that cannot be adequately captured by the variables we identified. In other words, 21st century law enforcement doesn’t solve crime by putting twelve people in a parlor; if there is forensic evidence at the scene, it gets analyzed, plonked into giant databases, and could generate hits that are one-in-a-million, not one in twelve.

Enter machine learning. As we’re all now figuring out through our use of ChatGPT, artificial intelligence excels at digesting large amounts of text, identifying repetitive patterns, and throwing those patterns into a model. AI is intertextual in that it can assess the impact of any factor in the database on any other factor. As my colleague Kristen Bell and others explain in this paper, this allows the tool to mine parole transcripts for repeated words to get a sense of factors that would not be salient to us in a traditional regression. Moreover, the capacity of these tools is enormous, so one can feed the machine tens of thousands of cases and get a very powerful sense of what is going on. There are even tools like SuperLearner, which can apply multiple machine learning tools to a dataset, coming up with the best of several models. My colleagues Ryan Copus and Hannah Laqueur do exactly this.

Machine learning has many applications in criminal justice, as this excellent NIJ article explains. The critiques that are leveled on machine learning often revolve around its most common criminal justice use: predicting reoffending risk. As explained in this solid blog post, critics worry that any predictive analysis based on historical crime data will reflect (and thus reinforce) existing biases embedded in the criminal justice system, and perpetuate misconceptions and fears through the feedback loop of basic predictions on past decisionmaking. In other words, as my colleague Sandy Mayson argues, the problem is with the nature of prediction itself. You rely on a biased past, you get a biased future.

What researchers like Bell, Copus, Laqueur and others contribute is the potential of turning the use of the predictive tool on itself and using it not to predict the risk of those subjective to the system, but rather the factors that impact the decisions that the system itself makes. For example, if private attorneys do a better job than state-funded panel attorneys, wouldn’t we want to know this, and wouldn’t it be important to figure out exactly what it is about their performance that makes the difference in the outcome? Using AI can help identify, for example, terminology used by lawyers, thus giving us a sense of the “flavor” of representation that parole candidates receive.

When done well, this technique has fantastic potential to teach us about the hidden nooks and crannies of the parole hearing machine that we would not be able to flag on our own. You don’t have to be an AI whiz to understand and appreciate machine learning research; you just have to understand what it does and appreciate its strengths and weaknesses.