Arnold writes: "I think that a bureaucracy gets established when a Bad Thing happens to an organization. The response is to set up the Department To Make Sure That Bad Thing Never Happens Again... I think that this explains why DEI bureaucracies... are so prevalent." Arnold goes on to note that it is not in the interest of the DEI bureaucracy to come up with a solution.
The situation is worse than that. They are set up to address a non-existent problem and so they have to create the appearance of one. They are pure grift operating under a specious moral cover.
Arnold quotes Cochrane: "from zero to about sixty thousand dollars of income, if you earn an extra dollar, you lose a dollar of benefits. Fix the incentives, and more people will get ahead in life." Is it even possible to fix this? Benefits without a cutoff become hugely expensive mass entitlements. Even a phase-out has the same disincentive effect. And in the bigger picture, the very large non-participation in the labor force suggests benefits may simply be too large, especially given the lack of work requirements to qualify. Earned income "tax credit," really a hand out since these people already don't owe income tax, is one approach, but not politically popular as anything other than an add-on benefit.
This unfortunate situation is not likely the result of inadvertence, but more likely deliberate. The more people kept in dependency the better it is for the bureaucracy and the political class. This is also likely a prime reason for open borders. The more people receiving benefits, the more bureaucrat administrators are needed; and the dependent are reliable votes for politicians promising to maintain and expand benefits. After the Clinton reforms which required showing up for work or training, welfare rolls dropped enormously, leading politicians to reverse these requirements as soon as they could.
We need more work requirements AND job offers AND individual tracking of gov't benefit recipients. And probably need gov't paid personal life coaches who get bonuses when the poor folk they coach stop being poor - stop making lifestyle choices which result in them being poor.
The Iron Law - orgs get captured by those who care more about the org than about the mission of the org.
Humans cooperate by forming orgs. When the org depends on serving customers, it gets good at that. When the org depends on getting allocated govt money by politicians, the org gets better at getting more money.
It's good for people to want to "do something" to solve problems. The idea of having "gov't do something" is mostly a long term mistake. We need ways of rewarding orgs who actually reduce poverty; reduce crime; reduce promiscuity; reduce illiteracy.
Freddie was able to talk over 30 minutes* about the mental illness issue in an admirable rephrasing of an article and some posts he's written recently. Since crazy Neely was threatening innocents and, in being restrained by an ex-Marine, was killed in the restraint.
Not explicitly discussed is the reality of both types of errors: (1) confining folks who don't need confining, and (2) not confining those who need to be confined. The pre-Cuckoo's Nest asylums had too many Type 1s, confining those who didn't need it. Today we have too many Type 2s, not confining crazies who need to be confined.
Freddie is correct that we need more involuntary confinement - yet his correct arguments aren't convincing most folk (like John Alcorn!). Freddie says (many of?) those folks are unthinking, and my skim reading of many such progressive folk's tweets seem to confirm that "unthinking" label/ evaluation/ derogatory remark.
What I'm missing is more clear criteria for when an mentally ill person, who acts ill but claims he is not, should be confined against his will. I assume that most dangerous folk are not always dangerous - which is one of the dangers.
Freddie: "To be severely mentally ill is to be held hostage by internal forces that are not you. This destroys every argument about personal freedom and choice, permanently and totally. Opponents of involuntary commitment never talk about this point or even attempt to rebut it because they know they can’t. "
John A: " Public safety might require coercion of some persons who really won't fit in. But call it what it is: coercion to prevent crime and/or to remove public nuisance. Don't cloak it in paternalism."
Both agree that coercion to prevent crime might be necessary. Often it is necessary to protect the public. It's also necessary to help the sick get a treatment to stop being sick which they would reject if given a choice, but their rejection is because they're sick. I wouldn't call this paternalism - but I also can't fully claim it's not, and certainly there is an argument that all treatment of dependents "for their own good" can be called paternalistic. Children need to be taken care of - so do many sick adults, without having legal age be the objective criteria.
The failure to treat the mentally sick hurts both the innocents hurt by the sick, as well as hurting the sick folk themselves. I'm pretty willing to leave alone those who do NOT defecate in public or leave drug syringes around, and who do NOT threaten innocent folk. But as small aggressions escalate into minor, or major, criminal behavior, the innocent public needs to be protected - and the sick person needs treatment.
Yet today many mentally sick folk are NOT getting cared for very well, and the public is stuck with a lot of nuisance, as well as crime. Of course, if theft <$1000 is not "real crime" (worth arrest & trial & jail), the crimes have to be pretty bad before the sick are coercively imprisoned in some health place. (Does health facility or health institution really convey more meaning, rather just signaling?)
It's now time to reduce the error of not confining those who need treatment - more than fear of over-confining. Those who object and claim we need less involuntary care are wrong, whether they don't know how hard it is to confine crazies or not. Tho their opinion is that I'm wrong - and in a case of conflicting opinions about the balance of errors, their claim is that too many or the right number are confined, while my claim is that not enough, like Jordan Neely, are confined.
Both I and those who disagree want fewer errors - but reality is such that reducing one type of error usually means increasing the other type. So then democracy becomes choosing to make a change to reduce one bad, while knowingly increasing another bad. If most political problems are like that now, this becomes a reason to increase ... mental illness in society.
*Since I had read the article and most recent posts, it was both interesting to see a public intellectual rehash his arguments, yet also disappointing that there was so little time to talk about Freddie's silly Marxism or his excellent acknowledgement of genetic influences on IQ and possibly other traits.
I haven't read Szasz, and would be interested to hear about whether he or others explicitly talk about both errors for treating mental illness.
"The most difficult problems aren’t budgetary or logistical. They are fundamental. They involve the involuntary nature of the care being called for and the flawed antipsychotic medications that are the mainstay of treatment for people dealing with the symptoms of psychosis"
"The mayor’s rhetoric refers to a psychiatric condition known as anosognosia — the state of being too sick, too far beyond reason, to recognize one’s own mental illness. It’s a diagnosis worth much debate, because it can be applied to anyone who doesn’t agree with a psychiatrist’s finding and can result in people being denied any real say in their own care. But it isn’t necessary to question anosognosia in order to question mandatory treatment. Because even if involuntary care may be warranted, the question remains: Does it work?"
"the World Health Organization has published guidance to eliminate involuntary psychiatric treatment altogether, because, according to Michelle Funk, who leads the W.H.O.’s work on mental-health policy, 'Involuntary treatment can harm a person’s mental and physical health, exacerbating crisis situations, damaging relationships with the clinicians, family members and others involved in coercive measures and driving people away from the mental health care system.'"
"Antipsychotics probably reduce hallucinations and delusions for around 60 percent of those who take them, but the science around their efficacy is far from definitive and some studies (though not all) indicate that long-term maintenance on the drugs may worsen outcomes.
Science hasn’t made great strides in antipsychotics since the drugs were first introduced seven decades ago. Their lack of precision remains largely the same, and because the drugs affect metabolic systems as well as dopamine pathways throughout the brain, they often have profound side effects: mental torpor, major weight gain, tics, spasms and a condition called akathisia, an overall jitteriness, as if a mad puppeteer is fighting perpetually for control of the person’s body.
Commonly, people abandon their antipsychotic drugs, whether they’re in mandatory treatment or the most sensitive, attentive voluntary programs. This is generally attributed to anosognosia and the disorganization that can come with mental illness, but it might well be seen as an outcome from the weighing of pros and cons."
"our current direction, toward more involuntary, medically centered care, probably won’t get us what we wish for: safer public spaces and fewer lost people.
We’re going to have to think less fearfully and more creatively, genuinely seeking the counsel of people who’ve learned to cope, in varied ways, with their psychiatric conditions."
Tom Grey's thoughtful, probing comment raises key issues. Let me address a few.
A) Re: "Freddie is correct that we need more involuntary confinement - yet his correct arguments aren't convincing most folk (like John Alcorn!)."
I hasten to clarify that *I don't know* whether current levels of involuntary confinement are excessive or insufficient.
Remember, DeBoer would impose involuntary confinement on two classes:
(1) Criminals
(2) Innocent persons, who are "severely mentally ill".
Consider criminals. For all I know, it might be the case that too many criminals get off the hook without incarceration; for example, via plea bargaining or lack of prison space. In that case, more involuntary confinement would make the world safer and more just. (Although more trials and more prisons might require great resources.) Or, instead, it might be the case that too many criminals are sentenced to incarceration, if there exist effective (but under-funded) alternatives to incarceration. It's an empirical issue.
Now, consider an innocent person who is "severely mentally ill". DeBoer would impose treatment and confinement on her, in the name of paternalism.
By contrast, my fundamental point is that innocent persons — including those who are classified as mentally ill — have liberty. A psychiatrist should not have power "to commit" an innocent person unless the person presents *a clear and present danger* to herself or to others. "Innocent" means not engaging in crime and not creating public nuisance.
Of course, there will always be a degree of law-enforcement discretion about enforcement-thresholds of crime (stealing an apple?) or nuisance (panhandling?). I don't pretend to calibrate the threshold in the abstract. Again, for all I know, perhaps enforcement-thresholds should be lower than they are currently.
Back to the innocent person. A diagnosis of anosognosia undermines the person's rights. (The psychiatrist finds the innocent person mentally ill, given her lifestyle, or because she has delusions, hallucinations, and problems of living, or because she creates social friction (really won't fit in). The innocent person rejects the diagnosis/treatment. The psychiatrist declares that the person's lack of deference and her non-compliance confirm the validity of the original diagnosis and justify coercive treatment/confinement.)
B) Re: Type I and Type II errors, and crime prevention.
Type I and Type II errors are concepts in statistics:
The goal of balancing Type I and Type II errors may help to guide various kinds of decisions in law enforcement; e.g., arrest, bail, discretionary sentences, and parole. A criminal trial verdict, which may impose a long period of incarceration, has a different formal decision criterion: Guilt must be established "beyond a reasonable doubt". The prosecution must persuasively counter every substantive point made by the defense. Some scholars interpret this criterion as a ratio of Type I and Type II errors: "It is better that ten guilty persons escape [Type II error, false negative] than that one innocent suffer [Type I error, false positive]." (Blackstone). Presumably, the optimal ratios in other decision contexts — arrest, bail, parole — would place much more importance on public safety than does Blackstone's ratio for trial verdicts.
Back to our discussion of involuntary confinement of innocent persons who have delusions, hallucinations, or other symptoms of "severe mental illness." We must not forcibly confine an innocent person in order to prevent crime. In my original comment, when I loosely used the phrase, "to prevent crime," I had in mind decision contexts like arrest on suspicion of crime/nuisance, bail, verdict, sentence, and parole; contexts in which your analysis of Type I and Type II errors is cogent.
In a liberal society, authorities mustn't forcibly confine a person *who isn't a suspect*, unless she otherwise constitutes *a clear and present danger*.
Note that the need to balance Type I and Type II errors is built into ethical decision-making about whether to arrest a person on suspicion of a crime/nuisance — and about whether to declare a person a clear and present danger.
If a person is not a suspect, then it is a violation of her rights — and surely would create a slippery slope — to impose preemptive confinement on her on the basis of (a) a diagnosis or (b) a subjective probability judgment (or even statistical evidence from sociology or psychology) about her *potential future criminal behavior*. A hypothetical future danger is not a clear and present danger.
If Durmstrang and Beauxbatons launched a "defence against whatever department" each, then Hogwarts would follow suit (probably poaching some experienced stuff from the other schools in the same time). That is, I feel that avoiding to fall behind in the latest managerial fad can be a goal itself for organisations, so the bar for replication is even lower than envisaged in the interesting musings discussed here.
"Roughly speaking, from zero to about sixty thousand dollars of income, if you earn an extra dollar, you lose a dollar of benefits. Fix the incentives, and more people will get ahead in life. We will also better help the truly needy, and the budget."
While I agree some face rather severe negative incentives, such as loss of Medicaid and that there are other lesser, and maybe greater, negative incentives to increasing one's income, I find the 1 to 1 ratio, even if accurate, to be rather unhelpful.
1 I'm really curious how it was estimated. How bad is the estimate? I'm guessing they didn't actually look at all the people getting benefits and calculate what they would lose with an extra dollar of income so how did they approximate this? Did they account for differences from state to state? What did they include as benefits? Home mortgage deduction? How did they account for benefits that are partly or wholly based on personal assets?
2 How much of these benefits go to people who successfully use the benefits to get out of poverty?
2a How much goes to people who will never be able to live independently due to some combination of physical and mental limitations?
2b How much goes to people in a temporary position needing help l?
3 I suspect some benefits go to people who would suffer severe negative incentives from increased income and some to people who would barely notice. This seems more important to me than the 1 to 1 ratio.
4 No doubt the policies for benefits are haphazard and sometimes (or often) miss the mark but how much can we reasonably expect the policies could be improved without more negative consequences (where the benefits were truly helping) than positive as a result of "Fix[ing] the incentives"?
I call straw man. DeBoer should reckon with Thomas Szasz, instead of focusing on whoever asserts that the homeless schizophrenic is "inherently harmless." There is plenty of room to disagree with DeBoer without being "unthinking" (Arnold's term).
Friction and conflict can readily arise between an individual who has major "problems of living" (Szasz's term) — who really won't fit in — and people who do fit in. Public safety might require coercion of some persons who really won't fit in. But call it what it is: coercion to prevent crime and/or to remove public nuisance. Don't cloak it in paternalism. Don't wrap it in the self-validating meta-diagnosis, by which a person's refusal to accept a diagnosis of "mental illness" confirms the diagnosis.
Schizophrenia is a physical brain disorder, no different in principle from any other brain condition such as epilepsy. Unless you think someone having a seizure on the ground is expressing their preferences for having seizures, you should reconsider your understanding of psychiatric illness.
To pursue your analogy: Should we confine an epileptic, against her will, on grounds of paternalism, because she is prone to involuntary seizures in public places?
If seizures are intrinsic to schizophrenia -- I'll defer to you on that -- the person nonetheless may have preferences about confinement, freedom, medication, and so on. To my mind, the criterion for *coercion* as a remedy is whether the person, who is labelled "mentally ill," causes substantial harms to others (nuisance, crime) -- not paternalism towards her, via self-validating diagnosis of anosognosia.
Compare Bryan Caplan's observations about Szasz on schizophrenia, in his article at the link below (2006, at pp. 336-40):
Seizures are not part of the schizophrenia spectrum. The point is that in schizophrenia, the symptoms such as delusions, hallucinations, disturbances in the ability to think logically, etc, are signs of a brain in massive distress and in the grips of a syndrome which distorts thinking, judgment, etc. It is an altered state. An epileptic seizure is also an altered brain state, although of course it manifests very differently than the symptoms of schizophrenia.
The individual doesn’t choose for any of this to happen and the psychotic states of schizophrenia are almost always unpleasant for them, or at the very least lead to significant morbidity and mortality. So yes, we are actually rescuing people from disordered brain states that are beyond their control.
And of course much of the rationale for involuntary commitment is the safety of others as well.
If the person does not engage in crime (or public nuisance), and if she
wants to be left alone, then we should not coerce or confine her. 'Save me from those who would save me from myself.'
Many people who wouldn't be labelled "mentally ill" have "delusions [fixed false beliefs] [...] and disturbances in the ability to think logically." On these criteria, few would be safe from coercive paternalism.
I grant that hallucinations [unreal perceptions] are the hard case. But I would err on the side of leaving her alone, if she wants to be left alone. Offer help or opportunity or positive incentives -- not coercion.
This is already pretty much the way it is. If someone isn’t either a danger to others or themselves, they are rarely committed, generally speaking. One of the big problems is that the resources and will to commit those who really do pose some danger are lacking.
If I understand correctly, Freddie DeBoer would commit, with coercive treatments, those who have "severe mental illness," against their will, in the name of paternalism and the self-validating diagnosis of anosognosia -- even those who aren't a menace to the public. Perhaps you and I would agree on policy, despite disagreement about Szasz. What prompted my initial comment is DeBoer's advocacy of coercive paternalism. I appreciate your thoughtful comments.
John, thanks for a good set of comments with Martin G. I've disagreed a bit based on Type 1 & Type 2 errors and the need to trade off between them in my longer comment that includes an excerpt of you.
Some of the sick folk need paternalistic care, for themselves.
Here (below) is what I wrote about incarceration, in comments to Arnold's blogpost on April 7. My point was that a subset of families are indeed harmed by incarceration of the father, and want him released. I don't know whether this subset is 10%, 30%, 60% or x% of incarcerated fathers. I noted that my point wasn't to advocate or to justify release. There is a proverb: "Justice is a harsh virtue." I would be happy to see, and to learn from, the data you offer.
My comment of April 7, re: incarceration:
"
John Alcorn
Apr 7
Re: "parental incarceration has beneficial effects on children" -- 2021 article in AER.
My intuitions:
(a) The statistical finding very probably masks a mix of positive and negative effects.
(b) A substantial subset of children very probably experience more harm than benefit, if the father is incarcerated.
I don't say this in order to argue against incarceration (incapacitation) of criminals.
My point is that condign punishment of a criminal, who has children (or spouse, or others who count on him or need him), intrinsically involves risk of major harm to intimate *innocent* third parties. This is a tragic aspect of a substantial fraction of real-existing punishment. Perhaps this is an implicit message in the old saw, "A good man is hard to find."
An aside:
Perhaps this is one of many reasons why Dante's masterpiece, *Inferno,* captures the imagination. In Dante's vision of Hell, the damned experience incarceration/incapacitation in a variety of characteristic punishments. Their fates are metaphors of the specific wrongdoings, and so communicate: "This is *how* what you did is wrong." More to the point, the metaphorical punishments -- unlike worldly incarceration -- don't impose negative externalities on the wrongdoer's innocent intimates, because the punishment occur in the afterlife. Presumably, infernal punishments have a positive externality -- general deterrence -- insofar as the vividness of Dante's poetry inspires deeper comprehension of wrongdoing, and greater fear of a purportedly ineluctable punishment in the afterlife.
[...]
[Then, in reply to others in same thread;]
Thanks for posting good sources. Yes, for many children, the quality of the father is more important than the presence. Incarceration of the father, then, is a blessing for the child.
My point (in this thread) is simply that another subset of children probably suffer more harm than good if the father is incarcerated. I guess it's a point about statistical generalizations. One wants a narrow, appropriate reference class, if possible."
I'm open to persuasion, to change my mind, if there are statistically representative data showing that few children are worse off when the parent is incarcerated.
Absent a clear dataset, I form my intuitions from casual observation, deference (thanks to friendships with a former Hartford PD head of the vice squad and then of the gang task force, and with a career prosecutor in Waterbury CT, first in family court, then in homicide division), fairly wide reading in criminology, and relevant works of narrative art. If that is "mood affiliation," I plead guilty.
To be sure, "substantial subset" is a very imprecise term. I don't know if the subset of children who are worse off is 10%, 40%, or x%.
I called attention to *heterogeneity* in impacts of parental incarceration on children. I acknowledged that a parent's incarceration can make children better off, when quality trumps presence. I didn't make any claims about the average effect of incarceration on children of offenders. I merely asserted that, in a substantial subset of cases, incarceration of a parent does more harm than good to the offender's children. (The point is empirical. I don't advocate any policy; e.g. "decarceration.")
I mentioned that my intuitions are based partly on reading in criminology. Here is a careful new study that address heterogeneity and finds that parental incarceration is bad for the offender's children in a specific, substantial subset of cases:
Will Dobbie et al., "The Intergenerational Effects of Parental Incarceration" (HKS and NBER):
"We estimate the causal effects of parental incarceration on children’s short- and long-run outcomes using administrative data from Sweden. Our empirical strategy exploits exogenous variation in parental incarceration from the random assignment of criminal defendants to judges with different incarceration tendencies. We find that the incarceration of a parent in childhood leads to a significant increase in teen crime and significant decreases in educational attainment and adult employment. The effects are concentrated among children from the most disadvantaged families, where criminal convictions increase by 10 percentage points, high school graduation decreases by 25 percentage points, and employment at age 25 decreases by 29 percentage points. In contrast, there are no detectable effects among children from more advantaged families."
A decade ago, a meta-analysis by Joseph Murray et al., published in Psychological Bulletin (APA), found that "the methodological quality of many studies was poor. More rigorous tests of the causal effects of parental incarceration are needed, using randomized designs and prospective longitudinal studies.":
"we draw together four of the largest studies to date on the effects of parental incarceration on children. These studies have followed children through their lives in four different countries: England, Sweden, the Netherlands, and the United States. [...] A critical issue for research is how to identify true causal effects of parental incarceration, over and above various background circumstances associated with incarceration, which also might propel children toward poor adult outcomes. This is a major challenge, and throughout this book, we discuss the concepts involved and methods used in trying to estimate causal effects on children's life outcomes. Tight causal conclusions will always be hard to achieve on this issue, but various methods of matching, statistical control, and analyzing change in children's behaviors before and after incarceration provide means to move us closer to this goal. [...] We consider how children might be affected differently according to the length of parental incarceration or the number of times that parents are incarcerated and whether mothers or fathers are incarcerated."
Arnold writes: "I think that a bureaucracy gets established when a Bad Thing happens to an organization. The response is to set up the Department To Make Sure That Bad Thing Never Happens Again... I think that this explains why DEI bureaucracies... are so prevalent." Arnold goes on to note that it is not in the interest of the DEI bureaucracy to come up with a solution.
The situation is worse than that. They are set up to address a non-existent problem and so they have to create the appearance of one. They are pure grift operating under a specious moral cover.
Arnold quotes Cochrane: "from zero to about sixty thousand dollars of income, if you earn an extra dollar, you lose a dollar of benefits. Fix the incentives, and more people will get ahead in life." Is it even possible to fix this? Benefits without a cutoff become hugely expensive mass entitlements. Even a phase-out has the same disincentive effect. And in the bigger picture, the very large non-participation in the labor force suggests benefits may simply be too large, especially given the lack of work requirements to qualify. Earned income "tax credit," really a hand out since these people already don't owe income tax, is one approach, but not politically popular as anything other than an add-on benefit.
This unfortunate situation is not likely the result of inadvertence, but more likely deliberate. The more people kept in dependency the better it is for the bureaucracy and the political class. This is also likely a prime reason for open borders. The more people receiving benefits, the more bureaucrat administrators are needed; and the dependent are reliable votes for politicians promising to maintain and expand benefits. After the Clinton reforms which required showing up for work or training, welfare rolls dropped enormously, leading politicians to reverse these requirements as soon as they could.
We need more work requirements AND job offers AND individual tracking of gov't benefit recipients. And probably need gov't paid personal life coaches who get bonuses when the poor folk they coach stop being poor - stop making lifestyle choices which result in them being poor.
The Iron Law - orgs get captured by those who care more about the org than about the mission of the org.
Humans cooperate by forming orgs. When the org depends on serving customers, it gets good at that. When the org depends on getting allocated govt money by politicians, the org gets better at getting more money.
It's good for people to want to "do something" to solve problems. The idea of having "gov't do something" is mostly a long term mistake. We need ways of rewarding orgs who actually reduce poverty; reduce crime; reduce promiscuity; reduce illiteracy.
Freddie was able to talk over 30 minutes* about the mental illness issue in an admirable rephrasing of an article and some posts he's written recently. Since crazy Neely was threatening innocents and, in being restrained by an ex-Marine, was killed in the restraint.
Not explicitly discussed is the reality of both types of errors: (1) confining folks who don't need confining, and (2) not confining those who need to be confined. The pre-Cuckoo's Nest asylums had too many Type 1s, confining those who didn't need it. Today we have too many Type 2s, not confining crazies who need to be confined.
Freddie is correct that we need more involuntary confinement - yet his correct arguments aren't convincing most folk (like John Alcorn!). Freddie says (many of?) those folks are unthinking, and my skim reading of many such progressive folk's tweets seem to confirm that "unthinking" label/ evaluation/ derogatory remark.
What I'm missing is more clear criteria for when an mentally ill person, who acts ill but claims he is not, should be confined against his will. I assume that most dangerous folk are not always dangerous - which is one of the dangers.
Freddie: "To be severely mentally ill is to be held hostage by internal forces that are not you. This destroys every argument about personal freedom and choice, permanently and totally. Opponents of involuntary commitment never talk about this point or even attempt to rebut it because they know they can’t. "
John A: " Public safety might require coercion of some persons who really won't fit in. But call it what it is: coercion to prevent crime and/or to remove public nuisance. Don't cloak it in paternalism."
Both agree that coercion to prevent crime might be necessary. Often it is necessary to protect the public. It's also necessary to help the sick get a treatment to stop being sick which they would reject if given a choice, but their rejection is because they're sick. I wouldn't call this paternalism - but I also can't fully claim it's not, and certainly there is an argument that all treatment of dependents "for their own good" can be called paternalistic. Children need to be taken care of - so do many sick adults, without having legal age be the objective criteria.
The failure to treat the mentally sick hurts both the innocents hurt by the sick, as well as hurting the sick folk themselves. I'm pretty willing to leave alone those who do NOT defecate in public or leave drug syringes around, and who do NOT threaten innocent folk. But as small aggressions escalate into minor, or major, criminal behavior, the innocent public needs to be protected - and the sick person needs treatment.
Yet today many mentally sick folk are NOT getting cared for very well, and the public is stuck with a lot of nuisance, as well as crime. Of course, if theft <$1000 is not "real crime" (worth arrest & trial & jail), the crimes have to be pretty bad before the sick are coercively imprisoned in some health place. (Does health facility or health institution really convey more meaning, rather just signaling?)
It's now time to reduce the error of not confining those who need treatment - more than fear of over-confining. Those who object and claim we need less involuntary care are wrong, whether they don't know how hard it is to confine crazies or not. Tho their opinion is that I'm wrong - and in a case of conflicting opinions about the balance of errors, their claim is that too many or the right number are confined, while my claim is that not enough, like Jordan Neely, are confined.
Both I and those who disagree want fewer errors - but reality is such that reducing one type of error usually means increasing the other type. So then democracy becomes choosing to make a change to reduce one bad, while knowingly increasing another bad. If most political problems are like that now, this becomes a reason to increase ... mental illness in society.
*Since I had read the article and most recent posts, it was both interesting to see a public intellectual rehash his arguments, yet also disappointing that there was so little time to talk about Freddie's silly Marxism or his excellent acknowledgement of genetic influences on IQ and possibly other traits.
I haven't read Szasz, and would be interested to hear about whether he or others explicitly talk about both errors for treating mental illness.
https://www.thedailybeast.com/jordan-neelys-life-could-have-been-saved-by-involuntary-treatment
Hello, Tom,
See Daniel Bergner's guest essay in the New York Times, "A Major Problem With Compulsory Mental Health Care Is the Medication" (June 2, 2023).
https://www.nytimes.com/2023/06/02/opinion/compulsory-mental-health-care-medication.html
Excerpts:
"The most difficult problems aren’t budgetary or logistical. They are fundamental. They involve the involuntary nature of the care being called for and the flawed antipsychotic medications that are the mainstay of treatment for people dealing with the symptoms of psychosis"
"The mayor’s rhetoric refers to a psychiatric condition known as anosognosia — the state of being too sick, too far beyond reason, to recognize one’s own mental illness. It’s a diagnosis worth much debate, because it can be applied to anyone who doesn’t agree with a psychiatrist’s finding and can result in people being denied any real say in their own care. But it isn’t necessary to question anosognosia in order to question mandatory treatment. Because even if involuntary care may be warranted, the question remains: Does it work?"
"the World Health Organization has published guidance to eliminate involuntary psychiatric treatment altogether, because, according to Michelle Funk, who leads the W.H.O.’s work on mental-health policy, 'Involuntary treatment can harm a person’s mental and physical health, exacerbating crisis situations, damaging relationships with the clinicians, family members and others involved in coercive measures and driving people away from the mental health care system.'"
"Antipsychotics probably reduce hallucinations and delusions for around 60 percent of those who take them, but the science around their efficacy is far from definitive and some studies (though not all) indicate that long-term maintenance on the drugs may worsen outcomes.
Science hasn’t made great strides in antipsychotics since the drugs were first introduced seven decades ago. Their lack of precision remains largely the same, and because the drugs affect metabolic systems as well as dopamine pathways throughout the brain, they often have profound side effects: mental torpor, major weight gain, tics, spasms and a condition called akathisia, an overall jitteriness, as if a mad puppeteer is fighting perpetually for control of the person’s body.
Commonly, people abandon their antipsychotic drugs, whether they’re in mandatory treatment or the most sensitive, attentive voluntary programs. This is generally attributed to anosognosia and the disorganization that can come with mental illness, but it might well be seen as an outcome from the weighing of pros and cons."
"our current direction, toward more involuntary, medically centered care, probably won’t get us what we wish for: safer public spaces and fewer lost people.
We’re going to have to think less fearfully and more creatively, genuinely seeking the counsel of people who’ve learned to cope, in varied ways, with their psychiatric conditions."
Tom Grey's thoughtful, probing comment raises key issues. Let me address a few.
A) Re: "Freddie is correct that we need more involuntary confinement - yet his correct arguments aren't convincing most folk (like John Alcorn!)."
I hasten to clarify that *I don't know* whether current levels of involuntary confinement are excessive or insufficient.
Remember, DeBoer would impose involuntary confinement on two classes:
(1) Criminals
(2) Innocent persons, who are "severely mentally ill".
Consider criminals. For all I know, it might be the case that too many criminals get off the hook without incarceration; for example, via plea bargaining or lack of prison space. In that case, more involuntary confinement would make the world safer and more just. (Although more trials and more prisons might require great resources.) Or, instead, it might be the case that too many criminals are sentenced to incarceration, if there exist effective (but under-funded) alternatives to incarceration. It's an empirical issue.
Now, consider an innocent person who is "severely mentally ill". DeBoer would impose treatment and confinement on her, in the name of paternalism.
By contrast, my fundamental point is that innocent persons — including those who are classified as mentally ill — have liberty. A psychiatrist should not have power "to commit" an innocent person unless the person presents *a clear and present danger* to herself or to others. "Innocent" means not engaging in crime and not creating public nuisance.
Of course, there will always be a degree of law-enforcement discretion about enforcement-thresholds of crime (stealing an apple?) or nuisance (panhandling?). I don't pretend to calibrate the threshold in the abstract. Again, for all I know, perhaps enforcement-thresholds should be lower than they are currently.
Back to the innocent person. A diagnosis of anosognosia undermines the person's rights. (The psychiatrist finds the innocent person mentally ill, given her lifestyle, or because she has delusions, hallucinations, and problems of living, or because she creates social friction (really won't fit in). The innocent person rejects the diagnosis/treatment. The psychiatrist declares that the person's lack of deference and her non-compliance confirm the validity of the original diagnosis and justify coercive treatment/confinement.)
B) Re: Type I and Type II errors, and crime prevention.
Type I and Type II errors are concepts in statistics:
https://en.wikipedia.org/wiki/Type_I_and_type_II_errors
The goal of balancing Type I and Type II errors may help to guide various kinds of decisions in law enforcement; e.g., arrest, bail, discretionary sentences, and parole. A criminal trial verdict, which may impose a long period of incarceration, has a different formal decision criterion: Guilt must be established "beyond a reasonable doubt". The prosecution must persuasively counter every substantive point made by the defense. Some scholars interpret this criterion as a ratio of Type I and Type II errors: "It is better that ten guilty persons escape [Type II error, false negative] than that one innocent suffer [Type I error, false positive]." (Blackstone). Presumably, the optimal ratios in other decision contexts — arrest, bail, parole — would place much more importance on public safety than does Blackstone's ratio for trial verdicts.
Back to our discussion of involuntary confinement of innocent persons who have delusions, hallucinations, or other symptoms of "severe mental illness." We must not forcibly confine an innocent person in order to prevent crime. In my original comment, when I loosely used the phrase, "to prevent crime," I had in mind decision contexts like arrest on suspicion of crime/nuisance, bail, verdict, sentence, and parole; contexts in which your analysis of Type I and Type II errors is cogent.
In a liberal society, authorities mustn't forcibly confine a person *who isn't a suspect*, unless she otherwise constitutes *a clear and present danger*.
Note that the need to balance Type I and Type II errors is built into ethical decision-making about whether to arrest a person on suspicion of a crime/nuisance — and about whether to declare a person a clear and present danger.
If a person is not a suspect, then it is a violation of her rights — and surely would create a slippery slope — to impose preemptive confinement on her on the basis of (a) a diagnosis or (b) a subjective probability judgment (or even statistical evidence from sociology or psychology) about her *potential future criminal behavior*. A hypothetical future danger is not a clear and present danger.
Liberty respects innocence.
PS: Compare Paul H. Robinson, "Punishing Dangerousness: Cloaking Preventive Detention as Criminal Justice," Harvard Law Review (2001):
https://scholarship.law.upenn.edu/faculty_scholarship/38/
Re: Bureaucracy replication
If Durmstrang and Beauxbatons launched a "defence against whatever department" each, then Hogwarts would follow suit (probably poaching some experienced stuff from the other schools in the same time). That is, I feel that avoiding to fall behind in the latest managerial fad can be a goal itself for organisations, so the bar for replication is even lower than envisaged in the interesting musings discussed here.
"Roughly speaking, from zero to about sixty thousand dollars of income, if you earn an extra dollar, you lose a dollar of benefits. Fix the incentives, and more people will get ahead in life. We will also better help the truly needy, and the budget."
While I agree some face rather severe negative incentives, such as loss of Medicaid and that there are other lesser, and maybe greater, negative incentives to increasing one's income, I find the 1 to 1 ratio, even if accurate, to be rather unhelpful.
1 I'm really curious how it was estimated. How bad is the estimate? I'm guessing they didn't actually look at all the people getting benefits and calculate what they would lose with an extra dollar of income so how did they approximate this? Did they account for differences from state to state? What did they include as benefits? Home mortgage deduction? How did they account for benefits that are partly or wholly based on personal assets?
2 How much of these benefits go to people who successfully use the benefits to get out of poverty?
2a How much goes to people who will never be able to live independently due to some combination of physical and mental limitations?
2b How much goes to people in a temporary position needing help l?
3 I suspect some benefits go to people who would suffer severe negative incentives from increased income and some to people who would barely notice. This seems more important to me than the 1 to 1 ratio.
4 No doubt the policies for benefits are haphazard and sometimes (or often) miss the mark but how much can we reasonably expect the policies could be improved without more negative consequences (where the benefits were truly helping) than positive as a result of "Fix[ing] the incentives"?
4a Might a UBI be part of the fix?
Re: Freddie DeBoer on mental illness.
I call straw man. DeBoer should reckon with Thomas Szasz, instead of focusing on whoever asserts that the homeless schizophrenic is "inherently harmless." There is plenty of room to disagree with DeBoer without being "unthinking" (Arnold's term).
Friction and conflict can readily arise between an individual who has major "problems of living" (Szasz's term) — who really won't fit in — and people who do fit in. Public safety might require coercion of some persons who really won't fit in. But call it what it is: coercion to prevent crime and/or to remove public nuisance. Don't cloak it in paternalism. Don't wrap it in the self-validating meta-diagnosis, by which a person's refusal to accept a diagnosis of "mental illness" confirms the diagnosis.
Schizophrenia is a physical brain disorder, no different in principle from any other brain condition such as epilepsy. Unless you think someone having a seizure on the ground is expressing their preferences for having seizures, you should reconsider your understanding of psychiatric illness.
Dr. Greenwald,
Thanks for your comment. I'll bite that bullet.
To pursue your analogy: Should we confine an epileptic, against her will, on grounds of paternalism, because she is prone to involuntary seizures in public places?
If seizures are intrinsic to schizophrenia -- I'll defer to you on that -- the person nonetheless may have preferences about confinement, freedom, medication, and so on. To my mind, the criterion for *coercion* as a remedy is whether the person, who is labelled "mentally ill," causes substantial harms to others (nuisance, crime) -- not paternalism towards her, via self-validating diagnosis of anosognosia.
Compare Bryan Caplan's observations about Szasz on schizophrenia, in his article at the link below (2006, at pp. 336-40):
https://econfaculty.gmu.edu/bcaplan/pdfs/szasz.pdf
See also a recent podcast interview of Caplan on Szasz, remarks on schizophrenia (cue times 53:00-60:00):
https://betonit.substack.com/p/szasz-podcast
Note: Caplan focusses on 2 other symptoms (not seizures): delusions and hallucinations.
Seizures are not part of the schizophrenia spectrum. The point is that in schizophrenia, the symptoms such as delusions, hallucinations, disturbances in the ability to think logically, etc, are signs of a brain in massive distress and in the grips of a syndrome which distorts thinking, judgment, etc. It is an altered state. An epileptic seizure is also an altered brain state, although of course it manifests very differently than the symptoms of schizophrenia.
The individual doesn’t choose for any of this to happen and the psychotic states of schizophrenia are almost always unpleasant for them, or at the very least lead to significant morbidity and mortality. So yes, we are actually rescuing people from disordered brain states that are beyond their control.
And of course much of the rationale for involuntary commitment is the safety of others as well.
If the person does not engage in crime (or public nuisance), and if she
wants to be left alone, then we should not coerce or confine her. 'Save me from those who would save me from myself.'
Many people who wouldn't be labelled "mentally ill" have "delusions [fixed false beliefs] [...] and disturbances in the ability to think logically." On these criteria, few would be safe from coercive paternalism.
I grant that hallucinations [unreal perceptions] are the hard case. But I would err on the side of leaving her alone, if she wants to be left alone. Offer help or opportunity or positive incentives -- not coercion.
This is already pretty much the way it is. If someone isn’t either a danger to others or themselves, they are rarely committed, generally speaking. One of the big problems is that the resources and will to commit those who really do pose some danger are lacking.
If I understand correctly, Freddie DeBoer would commit, with coercive treatments, those who have "severe mental illness," against their will, in the name of paternalism and the self-validating diagnosis of anosognosia -- even those who aren't a menace to the public. Perhaps you and I would agree on policy, despite disagreement about Szasz. What prompted my initial comment is DeBoer's advocacy of coercive paternalism. I appreciate your thoughtful comments.
John, thanks for a good set of comments with Martin G. I've disagreed a bit based on Type 1 & Type 2 errors and the need to trade off between them in my longer comment that includes an excerpt of you.
Some of the sick folk need paternalistic care, for themselves.
Tom, You got me thinking. I replied to your longer comment. Sorry about the delay -- a hectic day today.
Here (below) is what I wrote about incarceration, in comments to Arnold's blogpost on April 7. My point was that a subset of families are indeed harmed by incarceration of the father, and want him released. I don't know whether this subset is 10%, 30%, 60% or x% of incarcerated fathers. I noted that my point wasn't to advocate or to justify release. There is a proverb: "Justice is a harsh virtue." I would be happy to see, and to learn from, the data you offer.
My comment of April 7, re: incarceration:
"
John Alcorn
Apr 7
Re: "parental incarceration has beneficial effects on children" -- 2021 article in AER.
My intuitions:
(a) The statistical finding very probably masks a mix of positive and negative effects.
(b) A substantial subset of children very probably experience more harm than benefit, if the father is incarcerated.
I don't say this in order to argue against incarceration (incapacitation) of criminals.
My point is that condign punishment of a criminal, who has children (or spouse, or others who count on him or need him), intrinsically involves risk of major harm to intimate *innocent* third parties. This is a tragic aspect of a substantial fraction of real-existing punishment. Perhaps this is an implicit message in the old saw, "A good man is hard to find."
An aside:
Perhaps this is one of many reasons why Dante's masterpiece, *Inferno,* captures the imagination. In Dante's vision of Hell, the damned experience incarceration/incapacitation in a variety of characteristic punishments. Their fates are metaphors of the specific wrongdoings, and so communicate: "This is *how* what you did is wrong." More to the point, the metaphorical punishments -- unlike worldly incarceration -- don't impose negative externalities on the wrongdoer's innocent intimates, because the punishment occur in the afterlife. Presumably, infernal punishments have a positive externality -- general deterrence -- insofar as the vividness of Dante's poetry inspires deeper comprehension of wrongdoing, and greater fear of a purportedly ineluctable punishment in the afterlife.
[...]
[Then, in reply to others in same thread;]
Thanks for posting good sources. Yes, for many children, the quality of the father is more important than the presence. Incarceration of the father, then, is a blessing for the child.
My point (in this thread) is simply that another subset of children probably suffer more harm than good if the father is incarcerated. I guess it's a point about statistical generalizations. One wants a narrow, appropriate reference class, if possible."
Actually it's a "prior probability", which can be adjusted by data. It may well also be similar many other libertarians, but it's seldom "data free".
1 anecdote is not statistically significant, but it's infinitely more than 0.
I'll stand by my comment.
I'm open to persuasion, to change my mind, if there are statistically representative data showing that few children are worse off when the parent is incarcerated.
Absent a clear dataset, I form my intuitions from casual observation, deference (thanks to friendships with a former Hartford PD head of the vice squad and then of the gang task force, and with a career prosecutor in Waterbury CT, first in family court, then in homicide division), fairly wide reading in criminology, and relevant works of narrative art. If that is "mood affiliation," I plead guilty.
To be sure, "substantial subset" is a very imprecise term. I don't know if the subset of children who are worse off is 10%, 40%, or x%.
Am willing to follow good data and analysis.
I called attention to *heterogeneity* in impacts of parental incarceration on children. I acknowledged that a parent's incarceration can make children better off, when quality trumps presence. I didn't make any claims about the average effect of incarceration on children of offenders. I merely asserted that, in a substantial subset of cases, incarceration of a parent does more harm than good to the offender's children. (The point is empirical. I don't advocate any policy; e.g. "decarceration.")
I mentioned that my intuitions are based partly on reading in criminology. Here is a careful new study that address heterogeneity and finds that parental incarceration is bad for the offender's children in a specific, substantial subset of cases:
Will Dobbie et al., "The Intergenerational Effects of Parental Incarceration" (HKS and NBER):
https://www.hks.harvard.edu/publications/intergenerational-effects-parental-incarceration
Abstract:
"We estimate the causal effects of parental incarceration on children’s short- and long-run outcomes using administrative data from Sweden. Our empirical strategy exploits exogenous variation in parental incarceration from the random assignment of criminal defendants to judges with different incarceration tendencies. We find that the incarceration of a parent in childhood leads to a significant increase in teen crime and significant decreases in educational attainment and adult employment. The effects are concentrated among children from the most disadvantaged families, where criminal convictions increase by 10 percentage points, high school graduation decreases by 25 percentage points, and employment at age 25 decreases by 29 percentage points. In contrast, there are no detectable effects among children from more advantaged families."
A decade ago, a meta-analysis by Joseph Murray et al., published in Psychological Bulletin (APA), found that "the methodological quality of many studies was poor. More rigorous tests of the causal effects of parental incarceration are needed, using randomized designs and prospective longitudinal studies.":
https://pubmed.ncbi.nlm.nih.gov/22229730/
Compare Murray et al.'s book, which attempts causal analysis and addresses heterogeneity:
Murray et al., *Effects of Parental Incarceration on Children: Cross-National Comparative Studies* (American Psychological Association: Psychology, Crime, and Justice Series, 2014).
https://www.amazon.com/Effects-Parental-Incarceration-Children-Cross-National/dp/1433817438
Excerpts:
"we draw together four of the largest studies to date on the effects of parental incarceration on children. These studies have followed children through their lives in four different countries: England, Sweden, the Netherlands, and the United States. [...] A critical issue for research is how to identify true causal effects of parental incarceration, over and above various background circumstances associated with incarceration, which also might propel children toward poor adult outcomes. This is a major challenge, and throughout this book, we discuss the concepts involved and methods used in trying to estimate causal effects on children's life outcomes. Tight causal conclusions will always be hard to achieve on this issue, but various methods of matching, statistical control, and analyzing change in children's behaviors before and after incarceration provide means to move us closer to this goal. [...] We consider how children might be affected differently according to the length of parental incarceration or the number of times that parents are incarcerated and whether mothers or fathers are incarcerated."
Interesting links. The connection to conversion to Judaism is unclear to me
Antonio García Martínez -- the author Kling cites -- is a convert to Judaism.
Thank you! I somehow missed this.