In a column from late last year, the New York Times’s probed the contradictions between his commitment to philosophical liberalism and his growing unease with the pervasiveness and perniciousness of social media and other addictive products. In a paragraph that should ring true to regular readers of TCF, Klein writes:

[T]here are many kinds of products in which more innovation can lead to more destruction. Do we need vapes that are more compulsively usable? Is it good that online gambling firms are spending so much on slick marketing to find new users? Do we really want A.I. companies competing to create the most addictive pornbot? The question, I think, is under what conditions algorithmic media becomes such a product.

What Klein is talking about here is something I write about a lot—the rise of the modern, technologically enabled vice economy.1 Regular readers have heard this before, but as I put it in a recent National Affairs essay:

In 2023, a record 62 million Americans smoked pot; 17 million now use it daily or near daily. One in 12 young adults used a hallucinogen; one in 18 misused prescription stimulants such as Adderall. Another 2.6 million Americans over 12 took meth. Overdoses still claim the lives of 70,000 Americans annually; the majority died using synthetic opioids like fentanyl. Half of American men have a sports-betting account, up from almost zero seven years ago. "iGaming" — gambling via casino apps on your phone — is now legal in seven states. By some estimates, pornography now generates more revenue than Hollywood, and OnlyFans creators collectively make more than players in the National Basketball Association.

Contemporary American life is one big struggle with what the Greeks called akrasia, the phenomenon of doing something even though we know it would be better if we didn’t. And akrasia has long been a thorn in the side of liberalism—a fact with which Klein admirably struggles:

Modern liberalism is built around the idea that the government should make it possible for people to pursue their happiness as they see fit, so long as they are not harming others. It has much to say about individual rights and little to say about the common — or even the individual — good.

Liberalism carries, at its core, a trust that social experimentation will lead to better forms of social organization. That has freed it — and freed us — from the shackles of repressive traditions. But it can be confounded when adults are freely making decisions that don’t harm others but perhaps harm themselves. And it has created a loophole that algorithmic media companies have driven a truck through: We’re just giving people what they want, they say. Who are you to judge what they want?

Many kinds of liberalism are, to use the political philosophy term, anti-perfectionist. “Perfectionism,” for our purposes, is the idea that 1) there are ways of living that are objectively better and worse, and 2) the state should at least sometimes act so as to induce people to live well or not live badly, in line with that objective standard. To be a perfectionist does not mean thinking that the state should always and everywhere intervene in our choices—a panoply of reasons might justify non-intervention. But the perfectionist rejects the idea that the state should remain neutral about what the good life is, insisting instead that the good life should inform state action.

Many (although not all) liberals are, or at least profess to be,2 anti-perfectionists. Some liberals dissent from 1, arguing that we cannot make normative judgements about whether some ways of living are better than others. This is a view with more currency on college campuses and in far-left settings than in the mainstream left (although some still seems willing to defend it). Klein, specifically, seems to believe in at least some version of 1: people who are addicted to their phones are living worse lives, and it is meaningful to say so. And more generally, he seems to agree that there is such a thing as living well beyond “do what thou wilt.”

But where he stops short is 2:

[S]ensing that the present digital environment harms many of its users is a long way from knowing what would be better or what the government should do about it. As compelled as I am by the idea of bringing ideas of human flourishing back to the center of our politics, I turn queasy when I read the history of movements that have tried to do so. … The Progressive movement scored many victories, but there’s much in its history — from its embrace of phrenology and eugenics to forcing Native American children into boarding schools — that is repulsive. The past offers little succor to those who claim to know how to perfect, or even improve, the characters of others.

He does go on to carve out two areas where he thinks the state can act: children and business:3

But it feels to me like the outlines of an agenda — or at least ideas worth debating and trying — are coming more clearly into focus. Much of it revolves around two ideas: First, children should be more insulated from the ubiquity of digital temptations. Second, companies that want to shape so much human attention need to take on more responsibility, and liability, for what might go wrong.

I am happy to agree with the two poles of Klein’s policy program. If he and I can work together to protect children from vice and restrain business from commerce therein, that would be a great success.

But I do not think we can successfully form such a coalition—or, more precisely, can have a coherently anti-vice politics—without “bringing ideas of human flourishing back to the center of our politics.” It’s not enough to say that businesses shouldn’t sell something or children shouldn’t be exposed to it. For certain of the vices preying on Americans today, we need the state to say that they are hostile to human life,4 and proscribable as such.

And—I want to argue—our commitment to the idea that the state should remain “neutral” in the face of vice is part of how we ended up in the condition that Klein rightly recognizes as a problem. A state which does not understand its ambit as including some degree of making its citizens moral will inevitably make its citizens immoral. And the contemporary American state’s capitulation to anti-perfectionism—of either type 1 or type 2—is part of what has bred our current condition.

Moral Ecology

Everyone forgets how recent the new vice economy is. Turn the clock back to January 2016, and recreational marijuana is illegal in all but four states; sports gambling is illegal essentially everywhere; there’s no betting on your phone; OnlyFans is still months from launch; the fentanyl crisis has barely taken off; etc.

Much of this change is an unintended consequence, I have argued here and elsewhere, of technological progress. As technology has gotten better generally, we became able to produce higher quality vices, both legal and illegal, resulting in a more potent product and a more reinforcing experience.

Both this, and the sale of vice generally, is the result of the natural logic of the market. People want addictive goods, the legalizers correctly observe. If buyers exist, there will be sellers to meet them. If the sale is prohibited, then those sellers will be significantly suppressed—they will still do business, but in a more dysfunctional fashion. The effect of that prohibition is relatively large—as I never tire of reminding people, prohibition is why you can’t buy fentanyl at Wal-Mart. And the last decade’s worth of experience reinforces the fact that state control of vice has a significant impact on the intensity with which it is sold and consumed.

I lay all this out because the case against moral regulation is often predicated on an idea of state “neutrality” as to individual choice. Legalization is not endorsement, advocates of the former argue; it’s the state abstaining from taking a position, compared to prohibition’s affirmative “no.”

What has happened in the case of liberalization, though, is that removing formal legal restraint is in effect an affirmation of vice. Where the state doesn’t intervene, individuals can be and often are induced to consume addictive, harmful products. Either the state counterbalances this natural tendency, or it doesn’t. Whatever isn’t prohibited is permitted; whatever is permitted will be made available.

The market is not the only sphere in which to see this dynamic. There is a whole social environment in which our moral decisions are made. And state inaction shapes that environment as much as does state action; rescinding a prohibition has a social effect as much as implementing one. Again, there is no “neutrality”: state actions either make that environment conducive to a flourishing life, or to a vicious one.

Robbie George, in his 1993 book Making Men Moral, evokes the concept of ecology to capture the environment in which our moral decisions are made. He writes:

In neither case will the moral environment eliminate the possibility of moral goodness and badness, for people can be good in bad moral environments and bad in good moral environments. The point remains, however, that a good moral ecology benefits people by encouraging and supporting their efforts to be good; a bad moral ecology harms people by offering them opportunities and inducements to do things that are wicked. A physical environment marred by pollution jeopardizes people’s physical health; a social environment abounding in vice threatens their moral well-being and integrity.5

It is easy to see the recent spate of legalizations as a withdrawal of the state from any claim to shaping that ecology. Each of these legalizations has their own justification, of course—sports gambling was legalized because of the anti-commandeering provisions of the 10th Amendment, which have nothing to say about marijuana. But the common factor behind each of them is a current of anti-perfectionism. Yet the result has been a profound harm to the moral ecology—“neutrality” has yielded an effective affirmation.

The Amoral State is an Immoral State

After decades of culture war over moral issues, moreover, contemporary politics has been essentially evacuated of explicit moralism. The Democrats punted on the issue some time around the Clinton impeachment, and the Republicans followed in 2016 at the latest.

One of the effects of this phenomenon is that insofar as your politics is amoral, it becomes licit to compete politically on providing goods that were previously deemed out of bounds. I wrote about this phenomenon shortly after the 2024 election for the New York Times, observing that Harris and Trump had competed to capture valuable young male voters by catering to their desire for vice goods. Thus, Trump moved left on pot, while Harris tried to one up him by identifying marijuana industry jobs as part of her “opportunity agenda” for black men.

In a recent piece, Hill columnist Matt Lewis extended my observations into Trump’s second term, the agenda of which “includes a broader tolerance for industries that profit from addiction, distraction, and dopamine hits.” We see this in the administration’s looking the other way on the backdoor national legalization of gambling through prediction markets, and in Trump’s embrace of rescheduling. Not, of course, that the Democrats are likely to be any better; I have no doubt that the next Democratic president will go even further on the vice industries that they don’t dislike for political reasons.

I have occasionally jokingly referred to the phenomenon—politicians using vice to court votes—as the “Deadbeat Median Voter Hypothesis.” The basic insight is that people like vice because it’s fun and makes them feel good. Politicians always want an issue that unifies people, and which cuts across partisan lines. Promising people more vice does both. The thing that restrains us is perfectionism: the idea that vice is harmful to us, and that policy should be trying to steward our moral ecology. But if you drop that belief, then you don’t have a political problem.

Except of course you do have a political problem—it’s called demagogy, and everyone allegedly thinks it’s bad. Which is my point: when a democratic state attempts to be categorically neutral as to vice (i.e. anti-perfectionist), democratic politics—the politics of self-rule—decays, turning from rational self-governance to appeal to our base instincts.

I am not saying, of course, that our politics became corrupted merely in the past decade, or that our politics are corrupted only because of vice. But I am pointing to a single corrupting process in our politics, and arguing that this corruption has worked its way out from our leaders’ commitment not to be immoral but to be amoral.

The point generalizes. Vice is toxic to our moral ecology; in the absence of cultural and legal suppression, it flourishes like a weed. Simply saying that it is not the state’s business still tips the scales against people’s interest in living well, or at least in not living poorly in one of a series of well-recognized ways.

This is an argument against the type 2 anti-perfectionist: government’s choice not to regulate vice is not a “neutral” act. It is an affirmatively pro-vice decision. If you recognize that vice is a problem—if you accept proposition 1—then at least in the case of affirmatively deforming substances and practices (i.e. vices), state neutrality is not actually neutral at all, but rather facilitative of evil.

And this is true not merely at the level of policy, but at the level of posture. It matters that people like Klein be willing to make the argument that the state should, at least sometimes, be in the business of making moral judgements. Because if it isn’t, then for reasons of political economy as well as politics, it will be in the business of facilitating vice.

Can We Trust Government to Regulate Vice?

Some anti-perfectionists are willing to concede this point, but still oppose morals laws because “neutrality” is the lesser of two evils. Yes, they might say, the government not controlling vice will facilitate evil. But the evils done by the government acting to control vice are greater than those done by it doing nothing.

This is, more or less, what Klein does when he invokes the worst historical examples of morals regulation as a reason not to embrace the second premise of perfectionism. And he’s right: sometimes the government’s judgement about what is worse and better results in great evils. Grown women end up sterilized because they are judged “imbeciles,” for example. The argument in Klein’s case appears to be that the harms done by these errors are significant enough that the state should not judge at all, at least where individuals are concerned; we should not be hubristic enough to pass laws encouraging individual people to live well or not live poorly.

Of course, there are risks to not passing moral judgements. For much of our history, many Americans thought that it was good for one race to be enslaved by another. That was incorrect, and obviously produced great evil. The eventual solution was to impose a different moral judgement about what is good for people—namely, freedom from bondage—by armed force. People disagreed about this, and some wanted to resolve the dispute by dividing the union between slave and free, under the principle of “popular sovereignty.” Yet I doubt very much Klein would take Douglass’s side in that debate.

The standard retort here, of course, is that there is a salient difference between regulating actions which do harm to others (such as slavery) and those which merely do harm to the individual who “voluntarily” chooses the harm.6 By cleaving off force and consent from the rest of judgement about what is good for people, the anti-perfectionist liberal preserves the ability to say that people should be free to do what they want, so long as it doesn’t harm another person.

This returns to the point about the lesser of two evils. Yes, the anti-perfectionist might say, government could make people’s lives better. But the state is the monopoly on violence; violence is what it does; therefore any utterance by the state is an implied threat of acting violently. The only really morally justifiable or proportionate reason to use violence is to stop violence; therefore, we should only deploy violence when absolutely necessary.

At this point we have left behind somewhat Klein’s own stated position (which is that the government can interfere with businesses and children, but not adults). But I think the core is the same, which is a sense that the state cannot be trusted with enacting many kinds of moral laws.

I am in part sympathetic. There are obviously better and worse ways to enforce prohibitions, and I generally oppose kicking down people’s doors to ensure they are living well. But I think that descriptively, this is often not what moral lawmaking looks like. And this is not because of some contingent factor, but because the state is actually much less defined by the exercise of strength than the foregoing theory seems to presume. So while I am happy to agree that we should weigh the harms of moral lawmaking against the harms of vice, I want to suggest that in many cases, the balance is not so clear as the foregoing argument implies.

What Government Does

There is, after all, a great deal beyond stopping violence that we back up with the power of the state. The number of federal crimes (nevermind state) is allegedly uncountable; we all commit “three felonies a day,” in Harvey Silverglate’s memorable phrase. This fact, when trotted out, is usually meant to horrify you—you and your fellow Americans go through life constantly at risk of brutalization by the state, a reality that is always hammered home with the most gruesome of anecdotes.7

It is jarring to note that these fears mostly do not come to pass. Most crimes are not identified by the state, most identified crimes do not result in punishment, and the overwhelming majority that do are for the kind of crime that everyone agrees should be illegal. The parts of the state that make moral pronouncements are largely disconnected from the parts of the state that use coercive force.

What’s the evidence of this? Among state prisoners—themselves about 7 out of every 8 prisoners—malum in se crimes account for the overwhelming share of offenses. Roughly two in three state prisoners are in on violent offenses, for example. Just 5 percent are in on “other” public order offenses, a broad category covering “court offenses; commercialized vice, morals, and decency offenses; liquor law violations; probation and parole violations; and other public order offenses.” The pattern is much the same in federal custody. In other words, most people who are actually punished for crimes have committed a major offense, something most people agree the state should act against. Almost no one committed the kind of offenses that take up a great deal of Silverglate et al.’s attention.

That is assuming, furthermore, that they are caught in the first place. If you commit a murder in America, you have about a coin-flip’s chance of being apprehended. The outcomes get worse from there: robbery, rape, and arson report clearance rates of about one in four. Burglary is a one-in-seven proposition; steal a car, and you’ll get away with it more than nine times out of ten. It’s hard to measure, but the figures are almost certainly lower (based on arrest counts) for morals crimes. And that’s only counting clearances of the crimes that are actually reported to police, which are fewer than half of violent offenses and less than a third of property crimes (again, both figures likely higher than the clearance rate for vice offenses).

If, then, moral laws are primarily threats—commands with an associated risk of force if the command is not adhered to—then they are very, very bad ones. They are in most cases threats of such minimal significance that they do not deserve to be labeled threats. If I tell my five-year-old that if he does something bad, he will go to his room less than 1 percent of the time, he will proceed with abandon. Yes, I issued a threat, but it’s not of the heft to label it as such.

So we have a conundrum: in theory, the government uses its coercive strength to threaten people into doing many, many things other than not be violent. Yet in practice, the government actually carries out threats against a relatively small percentage of lawbreakers—never mind of the general public. If state utterances are threats, then the vast majority of them are empty ones. What do we make of this?

Many laws, of course, are meant to shape the behavior of the law-abiding. The real problem with over-criminalization—and it is a real problem, to be sure—is that it imposes onerous burdens on the great majority of citizens and businesses who dutifully follow the law. This is both intrinsically unfair and also deleterious to our national wealth—millions of hours spent complying with stupid laws have a real cost in addition to being a waste per se. (My points here about what law does generally should not be read as a defense of any one law in particular!)

But most of our many laws do shape behavior, for better or worse. And at least in principle, those laws exist because there is some reason we want to shape behavior in that way. Many of these reasons are allegedly amoral—public-health rules are ostensibly not moral laws, although why we say that I don’t understand. But even these “amoral” laws are about shaping behavior by some mechanic other than threat.

What I want to suggest is that when the state “speaks,” it is overwhelmingly not making a threat. When the state speaks, it is overwhelmingly issuing a moral judgement—a normative claim backed by an appeal to reason (however questionable), rather than an appeal to force. As a corollary, the potential harms of most laws against vice are not greater than the vices themselves, because the potential harms of most laws are factually quite low. Yes, a law can be enforced with undue violence. But in the vast majority of cases, they aren’t.

This is not merely an abstract claim about how the state ideally acts; it is a claim about how most states, throughout most of human history, have acted. This is because throughout most of human history, states were relatively weak. Limited technological reach and relatively low ratios of state to non-state power meant that while kings made many laws, their ability to enforce them were significantly limited by technological reality. Hammurabi’s Code simply could not be enforced in the way that modern laws are.

This is why policing as a practice—as distinct from the ancient night watchmen—only dates to the 19th century. Even then, the policeman was not responsible for enforcing the diversity of law so much as “keeping the peace”—for making sure that people were generally orderly in public. The government’s pronouncements on what ought and ought not to be done were largely orthogonal to what people mostly were jailed for.

It is not a coincidence that the earliest pangs of liberalism coincided with the rise of the centralized state and the development of modern bureaucracy—both of which are tools by which government utterances can be turned into force much more efficiently than was previously available. And liberalism does a great deal to help protect people against the abuses that that power can bring about.

But even today, the basic nature of law persists. Yes, the potential for abuse can be a good reason to prefer against a moral law. But it is not a dispositive reason, because law is in essence a moral utterance. And the potential for abuse is often not greater than the harms done by failing to make that utterance.

I am (as mentioned above) generally sympathetic to Klein’s program of affirmative policy addressing kids and businesses, rather than (for example) mass arresting people for using vicious goods. This is in part because I think use of vice is bad for the user, but does not make him evil, so he does not deserve punishment; and in part because I think it is quite easy for the costs of such extreme interventions to outweigh the benefits. Nor am I saying that all bad conduct should be prohibited; I am not advocating for banning lying, for example. I am talking specifically about vice as such.

But I am here arguing simply that in addition to Klein’s two pillars, the state should in the case of vice make the explicit statement that vice is bad, and it should often do so through the act of prohibition. These prohibitions are, historically, often minimally enforced—recall that basically no one was arrested for sports gambling even when it was illegal. Even alcohol prohibition was surprisingly under-enforced.

But the fact of the state drawing a moral line—of saying that such-and-such product is affirmatively destructive for its citizens, and should not be tolerated in public life—is fundamental to the maintenance of moral ecology. And we should not categorically reject that important function of the state for fear of abuse, because the risk of abuse is actually often not greater than the risk of doing nothing.

(Il)liberalism?

Some who have made it this far in this essay are going to accuse me of being an “illiberal”—a term much in vogue among both those who claim it and those who use it for abuse. To be fair, I have been criticizing core tenets of liberalism. But I am not interested in turning America into (an imagined version of) Hungary, nor in using the state to legislate submission to the church to which I do not belong.

Rather, I tend to think of myself as recognizing the need for somewhat “illiberal” means to liberal ends. That is to say, I am happy to endorse many of the virtues beloved of liberalism: tolerance, industriousness, a learned skepticism, etc. And I think the procedural and substantive liberties that are liberalism’s progeny are generally conducive to human flourishing—certainly more so than is tyranny.

But while I regard these as good, I do not take it for granted that they will obtain for humans or human societies. I do not believe that “social experimentation will lead to better forms of social organization”; people left to their own devices will just as often give in to their demons. As a result, the virtues of a free people must be formed by institutions, and the loss of those habits must be guarded against. In this sense, because I prefer the American tradition of liberalism (at least as originally conceived), I think that “ideas of human flourishing” have to be at “the center of our politics.”

If, moreover, you take seriously the reality of vice—if you reject type-1 anti-perfectionism—then I am not sure you can avoid reaching my position. If you believe that social media and pornography and weed and the addiction economy are degrading; and if you think that it is good to live free; and if you recognize that whether or not we live free is in part a result of our social and political context; then it is very hard to stop at the last possible moment and say “no no, the state should not tell people how to live.” Either the state sees virtue as part of its ambit, or we elect leaders who cater to our vices. There is no third option.

There are a lot of things I am not saying.8 I am not saying that the state should take an interest in every aspect of its citizens’ lives; that there is no room for privacy, or for moral agency. There is a great deal of room for both. And there are lots of non-moral reasons why the state should not make moral statements in any given domain. Because I believe in the American tradition of toleration, for example, I think the state should not prefer a religion; I think society is more peaceful and more orderly when it does not.

The point I am trying to make here is simply that the state has to make judgements on the issues that I care about—on the economy of vice—and it can’t escape doing so. And if you are the sort of person who takes seriously the consequences of the new vice economy, then you have to be willing to commit yourself to the state making at least some moral laws. Those moral laws do not need to be enacted with brutality—usually they aren’t—nor without due regard for all of the other things we hold dear as a liberal society. But if we think that vice is a problem, then it has to be okay for the state to say to it, loudly and clearly, “no.”

Subscribe now

Just over two years ago, I wrote a post on this Substack arguing that “America is Only Going to Get More Divided on Israel.” In it, I used data from Pew’s American Trends Panel (among other sources) to argue that the rising generation was much more divided on Israel than prior generations were, meaning that the then-nascent conflict over America’s relationship with Israel was only going to intensify.

Two years on, I feel comfortable saying that my prediction was basically on the mark, although I should perhaps have moved my timeline up. Only 46 percent of Americans now say they are sympathetic to the Israelis over the Palestinians, the lowest figure on record. The shares unfavorable toward the Israeli government and people have steadily grown. And while self-identified Conservatives continue to lean pro-Israel, those under 50 are evenly divided.1

This post is not about public opinion on Israel. It’s about public opinion on a group everyone closely associates with Israel: Jews. There appears to be strong conviction on the part of both Jews and those who dislike them that the war in Gaza and decline in public support for Israel have shifted American opinion on Jewish people more generally. Has America gotten significantly more antisemitic over the past five two years?

I don’t want to say that the answer is definitively “no,” because a loud and angry minority has done a great deal of harm. I suspect the intensity of antisemitism has gotten stronger in the past several years. But intensity is not the same as extent. And in fact, as this post demonstrates, what was true before is still true: most Americans are fine with Jewish people, or at least feel sufficient stigma around being antisemitic that they won’t admit it when you poll them. And that has, I think, substantial implications for recent debates around antisemitism and Jewish politics.

Subscribe now

To show this, below I present some analyses of data from the American National Election Studies, a quadrennial survey administered before and after each presidential election. One of the instruments ANES respondents are administered is a series of “feelings thermometers,” which ask respondents to say how “warm” or “cold” they feel toward a person or other object on a scale from 0 to 100. Specifically, participants are told:

Ratings between 50 degrees and 100 degrees mean that you feel favorable and warm toward the person. Ratings between 0 degrees and 50 degrees mean that you don’t feel favorable toward the person and that you don’t care too much for that person. You would rate the person at the 50 degree mark if you don’t feel particularly warm or cold toward the person.

In addition to asking about warmth toward people, the ANES also administers feelings thermometers for different groups, such as feminists, labor unions, big business—and Jews. I like these data for several reasons. One is that there’s actually (unsurprisingly) little polling about how people feel about Jews (rather than about Israel). Another is that the continuous scale hopefully attenuates (while not entirely obviating) social desirability bias—it’s easier to say you give Jews a 48 out of 100 than to say you don’t like Jews.

Figure 1, above, plots the share of ANES respondents in applicable waves who report that they feel cold toward Jews (i.e. who gave a response of less than 50). A few observations:

• There is a significant increase comparing the 2024 wave to the 2016 wave, and a barely significant increase comparing 2024 to 2020. Focusing just on the point estimates, the share of the population willing to say they feel cold toward Jews is at its highest level since 2004. Omitting that outlier year, it’s higher than at any point since 1976.

• In absolute terms, willingness to express coldness toward Jews is still quite rare. The actual rate is less than 6 percent in the 2024 sample.

• Recent coldness is about on par with levels in recent years, and notably below those of the 1970s.

In other words: there has been a real but small increase in measurable antisemitism in the period containing October 7, 2023. That increase is on the order of 2 percentage points. That may hide larger real changes, although it’s not obvious to me why the social-desirability adjusted measure would shift more slowly than the real value.

Asking what fraction of respondents are “cold” toward Jews doesn’t give us a full picture of the distribution of responses to the thermometer question. Figure 2 (above) does, allowing us to compare how that distribution changed between 2020 and 2024 (before and after October 7, i.e.). It suggests that there has been some minor shifting in the top half of the distribution — fewer people giving a response between 91 and 100 and 61 to 70, and more giving responses in the 51 to 60/71 to 80 range.

It also shows that most of the growth in coldness comes at the very bottom of the distribution. To break it out, about 4 percent of respondents gave a score of less than 40 in 2020, verus about 5.6 percent in 2024. Most of that change comes in the 1-10 category, which grew by about a percentage point between those two periods.

One interesting question these data can answer is how warm people feel toward Jews *relative to* other groups. Relative disposition can be informative: if someone says they are 50 on Jews but 90 on all other groups, we might still think that they have some unusual views about the former.

The 2020 and 2024 ANES include thermometers on a number of other identity groups.2 To construct Figure 3, I take the average of each respondent’s warmth towards those groups, then subtract it from the respondent’s warmth toward Jews. If the resultant value is negative, they’re warmer toward other groups on average than they are toward Jews; if it’s positive, they’re warmer toward Jews than they are toward other groups on Average.

On eyeballing, Figure 3 seems to suggest that Americans are mostly equally warm or warmer toward Jews than they are toward other groups on average. That’s not uniformly true, of course; some are a bit colder, and a few are much colder. But scores are denser on the right-hand side of the distribution, in general, suggesting that many more are warmer than are colder. In other words, more Americans are warmer toward Jews than other groups than are colder toward Jews than other groups.3

The other notable takeaway is that people have gotten, if anything, relatively warmer toward Jews since 2020 (i.e. the right half of the distribution is denser in the 2024 data than in the 2020 data). That suggests the decline in warmth toward Jews since 2020 may represent a decline in warmth generally.

Perhaps these estimates are eliding some change in the population composition that might be more alarming. We tend to think that some people are more likely to be antisemitic than others. Are, for example, the rising generations less philosemitic than predecessors? (This comes up in my Israel opinion post.)

Figure 4 presents the results of a series of bivariate linear probability models4 for a few such variables. In each case, the DV is the probability that someone reports < 50 on the Jewish thermometer question. Note that these are all relative measures: because these are all categorical variables, each has a reference group. So, for example, being in the 68 to 80 age bracket reduces your probability of being cold toward Jews by about 4.2 percentage points. (Keep in mind that coldness is overall quite rare: again, about 6 percent of respondents (± 1 point) were cold in 2024.)

Notably, none of the political view options are significant: there’s some variation, but in general it seems like political view does not have a major effect on coldness toward Jews. (Interestingly, the two views that are coldest toward Jews are extreme liberals and moderates.) On the other hand, the demographic variables do seem to make a big difference: foreign-birth, birth of foreign-born parents, race other than non-Hispanic white, and lower age all seem to increase the odds (still quite small in absolute terms) of reporting coldness toward Jews.

Figure 5 is a multivariate regression trying to disentangle which of the demographic factors does the driving. (Please note that it’s a logit model, not a linear probability model—the coefficients are not interpreted the same way.)5 It seems like the answer is basically a) being foreign-born or born of foreign parents or b) being black. (Also being old is protective). This is just three variables, though, so I wouldn’t quote it as gospel—e.g. there’s nothing on SES in here. But it is interesting.

Subscribe now

In some senses, you could use all these data to corroborate a narrative increasingly common among people who spend a lot of time thinking about Jews — both those who like them and those who don’t. In this theory antisemitism (as measured by coldness) is increasingly common (almost as common as it was 50 years ago), as indicated both by these data and the sharp uptick visible online and on campuses. This effect is driven particularly by the young (and, in some tellings, recent arrivals), which means it will only get worse over time. The surveys are probably understating the true magnitude of the problem.6 The future of America is going to be antisemitic. If you like the Jews, this probably alarms you; if you don’t like the Jews, you look at these developments with glee.

I think this argument is not totally without merit. There does appear to be a change in the level of antisemitism in the ANES, directionally consistent with the change in the level of support for Israel (see start of post). And it’s not unreasonable to be worried about the former.

But I also think these data are consistent with—are arguably much more consistent with—another, simpler claim: Americans are basically fine with Jews. Why do I think this?

• Explicit coldness toward Jews is and remains extremely rare. In a 1962 survey, “between 17 and 25 percent of Americans thought that Jews had too much power [and] between 28 and 38 percent of Americans would consider voting for an antisemitic candidate.” These figures were down from the Great Depression. If you fit a trend line to antisemitism over the past hundred years, it slopes straight down; a several point bump amid a major ongoing war in the Jewish state is not anything like the magnitude needed to return to the bad old days.

• Extremely high warmth toward Jews remains even more common than coldness. In the 2024 ANES, about 6 percent of respondents were cold toward Jews; 21.2 percent (3.5 times as many) gave Jews a warmth score between 90 and 100.

• It is possible that the recent decline in warmth is driven in part by a decline in warmth generally. If you model the relationship between my general warmth variable and warmth toward Jews in the 2024 ANES, each one point increase in general warmth increases warmth toward Jews by 0.91 points. That increasing hostility toward Jews goes with increasing hostility toward other groups is not necessarily comforting. But it suggests there may not be a unique problem.

• Even if we assume the demographic variables represent some close approximation of their causal effect (and they don’t), the actual effects are small. In the bivariate nativity model, foreign birth increases probability of coldness from a baseline of 4.6 percent to 10.3 percent. In other words, you should still expect about 90 percent of foreign-born respondents to the ANES to give Jews a 50 or higher on the thermometer.

In other words, yes, there has plausibly been a level change in America’s views of Jews (not just Israel). But the level such as it is is still extremely favorable to Jews. There are, I think, two implications to this.

Subscribe now

Implication the first: Do not expect antisemitism to be electorally popular.

I’m not going to rehearse the whole saga with Tucker Carlson, Nick Fuentes, and Kevin Roberts.7 But one interpretation of events is that there’s an ascendant coalition of Jew haters on the right who will determine the future trajectory of the GOP. Rod Dreher (not exactly known for his temperateness) has repeatedly asserted that “between 30 and 40 percent of the Zoomers who work in official Republican Washington are fans of Nick Fuentes.”

I have factual beef with Dreher’s take here. I and lots of other conservatives—people who live in America, not Hungary—think this number is made up. People routinely overestimate the size of small groups, and I suspect that when Dreher claims that 2 out of every 5 young conservative staffers are Fuentes fans, it’s because he and his interlocutors are bad at mathematic intuition.

Which is not to say, of course, that there’s no antisemitism problem on the young right. There is, quite obviously; my Twitter mentions testify as much. I don’t think I should have to tell you that I am against it, but of course I am. Separately, though, I am concerned about a dynamic in which the rest of the right convinces itself that it needs to make peace with these people, or at least tolerate them, for the sake of coalitional interests and the future of the GOP. I think, as others have argued, that this is not only morally wrong but also extremely bad politics.

And it is extremely bad politics because most Americans are fine with Jews. It is electoral poison to be on the 30 percent side of a 70/30 issue; open Jew bashing is being on the 5 percent side of a 95/5 issue. Treating these freaks with respect and insisting they need to be part of the conversation is a bad thing to do, but it is also a great way to ensure that you lose. The Democrats in many ways made this mistake with their own lunatic fringe five years ago; the object lesson there is that you shouldn’t hand power to people with ideas that most of the public hates.

This brings me to the second implication: if most people are not Jew haters, you should probably be pretty careful about whom you call a Jew hater.

When you take an introductory statistics class, and you first learn Bayes Rule, you often go through an exercise to teach intuitions about false positives. Imagine an extremely rare disease, and a test that identifies the disease correctly 90%+ of the time. If the disease is sufficiently rare, the math works out such that a substantial fraction of positives will be false—you shouldn’t actually trust your test, and should always double check.

I find this exercise useful for thinking about accusations of antisemitism, which come almost as fast and furious on X as do the weird antisemitic posts. If, in fact, only about 5 percent of the population is openly antisemitic (and that’s probably an overstatement), then a test that has even somewhat frequent false positives will often get it wrong. Or, to translate: if you accuse a lot of people of being an antisemite, and most people aren’t antisemites, then many of your accusations are going to be wrong.

Maybe this is not a big deal on its own — tests are wrong all the time, we just run them again. But, actually, accusing people of bigotry is kind of a big deal, especially when those people are not bigots! And (I will go so far as to say), a great way for Jews to increase people’s dislike for us is to spend our time accusing our friends of being our enemies. Of course, people are sometimes actual antisemites. But they are with sufficient rareness that caution is warranted in making such accusations.

That is to say: if you recognize that most Americans are fine with Jews, then you certainly shouldn’t spend time trying to make friends with the ones who aren’t. But you also shouldn’t treat people like they’re something that they’re not—that’s a great way to lose friends.

Share

Mayoral Livestream

City Journal Substack

Zohran Mamdani Wants Civilians to Replace Cops. Will It Work?

Zohran Mamdani, New York City’s likely next mayor, has a complicated history with the police. Once an enthusiastic supporter of “defund,” Mamdani has tempered his past radical views during the campaign, going so far as to issue a general apology to the NYPD…

Read more

5 months ago · 2 likes · 1 comment · Charles Fain Lehman

I won’t spoil the whole piece, but it’s probably not what you’re thinking! Check it out (and please subscribe.)

As a result, we often reduce the problems associated with homelessness to the question of how to get people a permanent home (in much the same way that we assume poor people’s problems stem from a lack of money). This is one way of explaining the “Housing First” theory of homelessness policy, which stipulates that people should be given permanent housing as soon as possible, regardless of their drug use, mental health, or other problems. HF advocates often look askance at more temporary or indirect solutions, which they see as getting in the way of getting people housed. (This is classic “root causes” thinking.)

Often absent from this debate is discussion of the social externalities of homelessness—public disorder, crime, all the topics I like to write about here. Sometimes these issues aren’t acknowledged, and those who insist on talking about them are tarred with various derogations. When they are acknowledged, the solution is often deemed to be getting people into permanent housing—i.e., making them not homeless—thus returning the discussion to the fundamental American problem of too little housing.

A new working paper from Stanford’s Derek Christopher, Mark Duggan, and Olivia Martin offers a rigorous intervention into this debate. It both demonstrates that unsheltered homelessness has real social costs, and challenges the idea that those costs are best, or only, remediable by getting people into permanent housing. It turns out that simply getting people into temporary shelter—often derided by the homelessness-industrial complex as a band-aid—significantly reduces both social and individual harms of unsheltered homelessness. Those harms, in other words, can be addressed without providing people with permanent housing—meaning that we don’t need to wait for the eternity that new housing requires in order to address them.

Subscribe now

Christopher et al. focus on the effect of temporary shelter. Temporary shelter’s efficacy is, of course, often contrasted with that of permanent housing. As the authors note, “[p]roponents argue that [shelter] serves as a cost effective public good that mitigates the social costs of unsheltered homelessness, while critics assert that it diverts scarce resources from permanent housing solutions that more effectively reduce total homelessness.”

What are the social impacts of opening new shelters? To answer this question, the authors collected data from Los Angeles County’s homelessness system, crime data from the LAPD/LASD, and ER data from 2014 to 2019. They also use data on 170,000 unique individuals across 330,000 services check-ins to measure individual-level effects.

Why collect all of this data? In part, it’s because L.A. is a good context for studying homelessness: Los Angeles County is home to 3 percent of the nation’s population but 10 percent of its homeless population. But it’s also because of how the county approaches its problems. Every winter, it launches an expanded shelter program, temporarily adding 1,000 to 1,500 beds. The exact number of beds, opening and closing dates, and site locations all vary year to year. That creates exogenous variation that the paper exploits to estimate the causal effects of an additional shelter bed.

First stage: does the temporary expansion of shelter actually result in more people in shelter? Yes, the authors find—an additional 100 beds result in 89 additional people in shelter, “contradicting theories about widespread resistance to shelter among people experiencing homelessness.” They also find no corresponding increase in total homelessness, meaning that an increase in sheltered homelessness corresponds to a decrease in unsheltered homelessness.

What does moving people off the streets in turn do? One thing it does is reduce crime—an additional 100 shelter beds prevent an average of 1 crime per day. These effects are driven by violent crime (as above) and are strongest at night, when the emergency shelters are open. The authors note that the effects could be driven by reductions in either victimizing or victimization—the homeless are often the targets of violent crimes, and giving them a safe place to sleep may reduce these offenses.

“At first glance, this [effect] may appear small,” the authors write. “However, during this period, LAHSA’s winter shelter program operates around 1,500 beds per day for roughly 4 months per year. So, in total, the program prevents nearly 15 crimes every day it operates or more than 1,500 crime incidents every year.”

What about serious mental illness? The estimates imply that every 400 additional beds mean one fewer daily ER visit for psychiatric symptoms. Because temporary shelters don’t usually offer psych services on-site, this is likely the result of sleeping on the street being harmful to the mental health of occupants, and shelter mitigating that effect. The paper also finds suggestive evidence of a reduction in injuries and poisonings, “which are frequently referred to as the second most common reason for ER visits among homeless populations.”1

In addition to these estimates, the authors use data on individual homeless people to estimate the person-level effects of exposure to shelter. They find no effect of shelter availability on later usage of services, but there is evidence of a reduction in observed mortality (though that may be confounded by the process by which the data are observed.)

In other words: getting people into shelter reduces crime, ER usage for psychiatric issues, and (possibly) homeless mortality. As the authors put it, “temporary shelter functions as a high-value public good that generates substantial social benefits despite not ‘solving’ homelessness through permanent exits.”

Subscribe now

There’s an interesting juxtaposition here, I should add, to the research on Housing First. Housing First does mechanically increase the likelihood of having a home: people who get Housing First stay housed longer. Conversely, however, Housing First has no effect on criminal offending or mental health outcomes relative to treatment-as-usual. Treatment-as-usual, in this case, is usually shelter, combined with whatever other patchwork interventions the jurisdiction supplies.

Read this in the context of Christopher et al., and the reasonable conclusion is that many of the problems associated with homelessness are the result of unsheltered homelessness. Give people somewhere to go—shelter or permanent housing—and the problem gets better. Having a lot of people on the street makes many things worse.

The argument then becomes entirely about cost. Unfortunately, today it costs a lot to build affordable housing. We can wait around for publicly funded permanent housing to be built, but it seems like it doesn’t actually do much to address the costs of homelessness. At the margin, shelter almost certainly buys you more bang for your buck.

This last point brings me back to how we think about homelessness and homelessness policy. Often the question we ask—including in the HF debate—is “how can we solve homelessness?” This is the motivation beneath expansive promises to end homelessness. We have spent tens of billions of dollars toward this end, with the result being that homelessness is about as bad as it’s ever been.2

But it turns out that you don’t need to “end homelessness” to address many of the serious social problems associated with unsheltered homelessness. You just need to get people off the streets. When you do, dysfunction goes down, even if you don’t resolve the “root causes” of the problem.

• I’m in the Atlantic this morning with a piece on why the perception of safety and order matters to the success of public projects, especially public transit. That piece started life as sort of “abundance” adjacent, so if you’ve come here for my writings on that topic, you may enjoy it.

• City Journal, where I am senior editor and a regular contributor, now has a Substack. My first piece for it, about why altruism is a bad solution to the housing crisis, is live as of yesterday. I’ll be writing every other week on Wednesdays over there, so if you want more from me, I encourage you to subscribe, subscribe, subscribe!

City Journal Substack

City Journal is the nation’s premier urban-policy magazine. Produced by the Manhattan Institute for Policy Research Inc.

By Manhattan Institute

I had a new piece over at The Dispatch on Monday about the law and ethics of AI-generated child pornography/child sex abuse material (CSAM). That piece was prompted by a tête-à-tête on X between me and the usual “just asking questions” defenders of the manufacture of artificial CSAM.

I encourage you to read the whole piece, but the thrust is that AI CSAM currently exists in a legal gray area thanks to a 2002 Supreme Court ruling. In that case, the Court ruled that AI CSAM was, like most pornographic images, speech; that it was neither real CSAM (which is unprotected speech under Ferber) nor always obscenity (which is also unprotected under Miller); and that consequently, it couldn’t be proscribed simply as AI-CSAM, but only if a given instance falls into one of those two categories. I argue that this reflects our cultural inability to think about kinds of speech which are wrong in themselves, and our unwillingness to say that such speech should be proscribable. And I argue that Ashcroft v. Free Speech Coalition should therefore be overturned.

Since that piece came out, Sam Altman has announced that soon, ChatGPT and other OpenAI products will permit “erotica for verified adults.” Obviously, I do not think that OpenAI will start distributing Ashcroft-approved AI CSAM to anyone who puts in the right query.1 But I do think this opens up a rather larger question about the problem of AI-enabled obscenity and its socially disruptive effects.

After all, my argument in the Dispatch is not actually, to use a technical term, correct. (I would prefer to say that I oversimplified for purposes of getting the point across). To understand what I mean, you should know that under federal law, there are actually two separate criminalizations that could apply to AI-CSAM while surviving judicial review. One, 18 USC 2252A, prohibits AI-CSAM images which are “morphed,” i.e. which depict real children whose images have been altered to appear lewd, or which are produced from training data that contain real CSAM.

The other is 18 USC 1466A, which proscribes the production and distribution of “visual depictions” of children engaged in “sexually explicit conduct” which is also “obscene.” To the extent that there is any AI-CSAM not covered by either of these statutes (read: protected by the First Amendment), that space is defined largely by what, exactly, constitutes an “obscene” depiction of children engaging in “sexually explicit conduct.” That is to say: to the extent that AI-CSAM can’t be proscribed, it’s entirely because juries or courts are unwilling to call it obscene—the extent of obscenity is the extent of the prohibition.

This is, I never tire of observing when writing about this topic, obviously intrinsically absurd. Any depiction of children engaging in sexually explicit conduct is obscene in the conventional sense of the term—how could it not be? Yet several appeals courts have insisted that it’s possible, and found instances in which it exists. As I argued in the Dispatch piece, AI-CSAM can be optimized to fit into this niche; generating images just shy of obscene is a task at which AI will excel.

The reason the courts think non-obscene AI-CSAM exists is that obscenity has a technical legal definition under Miller v. California. For speech to be obscene (and therefore not receive the protection of the First Amendment), it must meet the following three requirements:

• the average person applying contemporary community standards would find the work, taken as a whole, appeals to the prurient interest;

• the work depicts or describes, in a patently offensive way, sexual conduct specifically defined by the applicable state law; and

• the work, taken as a whole, lacks serious literary, artistic, political or scientific value.

It’s conceivable that AI-CSAM may not be “patently offensive,” or that a jury may be persuaded that it has “serious … value.” That’s how you get to protected AI-CSAM.

If this definition seems extremely narrow, you are correct—it is deliberately so. The Miller standard is maximally speech-protective, both for principled reasons—speech is a right, and rights should be secured—and for contingent ones—a lot of the fight about obscenity in the ‘50s–‘70s was about overly censorious communities trying to proscribe obviously unobjectionable content. It is not, as I detail in that Dispatch piece and elsewhere on this Substack, consistent with what the First Amendment meant prior to the 1960s and 1970s.

And, as I have argued (again, at the Dispatch), the high standard for obscenity combined with the lax enforcement of obscenity laws (still very much on the books, thank you!) has left the state totally powerless to check the rise of hardcore pornography. As I’ve written previously here at TCF, the unregulated market in porn takes a predictable course, with more and more hardcore content being delivered in more and more direct ways. Much as in the case of incitement, we have removed the government’s long-standing ability to deal with a well-established social problem that happens to involve speech; as a result, that problem has become much more prevalent.

With incitement, of course, the problem is narrowness of definition. That’s a problem with obscenity, too, but it’s not the only problem. The problem is really one of deference—what the Miller court did was try to take away from the local community (embodied in the jury) the ability to judge for itself when something is and is not obscene, and replace that judgment with a strict standard that a jury was to algorithmically apply.

There is, of course, a lot of merit to this approach to the criminal law generally. But in the case of obscenity, it is impossible, because obscenity is not an objective thing. It is, rather, a severe departure from the community’s standards about what ought and ought not to be said—or depicted—about sex. Those norms fluctuate, and have always done so (there’s a lot of very dumb legal writing on this topic, as though observing norms change over time is some earth-shattering insight). The proscription of obscenity acknowledges, however, that all communities have an interest in regulating sex and the representation thereof—that what happens in the bedroom does not always stay in the bedroom, particularly not in the age when depictions of someone’s bedroom are in everyone’s pocket.

Which brings me back to OpenAI and the dissemination of AI obscenity. Because that’s what Altman is promising, of course: the manufacture of obscene materials at scale. I am sure that his models will be studiously tuned to avoid the legal definition of obscenity.2 (As I noted in the Dispatch, even Pornhub’s terms of service prohibit posting obscenity!) But in the colloquial sense of obscenity, that is what OpenAI will be facilitating.

I think this will be extraordinarily socially deleterious. I think it will turbocharge the constitutionally thorny3 question of revenge porn — if distributing your ex’s dirty photos might be protected speech, morphing him or her into dirty positions definitely is. I think it will raise further the opportunity cost on dating and mating already imposed by pornography. I think it will contribute further to the breakdown of relations between the sexes that the rise of widely available hardcore pornography has obviously abetted.

And most importantly, I think we have the tools we need to stop it. We have the tools we need to stop AI-CSAM, of course: just define it all as obscene, categorically, and move on to the next question. But we have the power, too, to say that it is wrong for OpenAI to serve a smorgasbord of infinitely customizable pornography to anyone who types in the right terms. We can subject that content to our collective judgments, through the power of the law. And OpenAI—a big company that would like to get even bigger—would studiously respect that power, if we chose to exercise it.

But we don’t, so it won’t. Nothing about this is inevitable: we just have to be willing to say that we know it when we see it, and we don’t like what we see one bit.

Subscribe now

The piece also dovetails nicely with my appearance on the House of Strauss podcast, which I encourage you all to check out.

With football season creeping closer, sports broadcasters are preparing for record attention to the national pastime. Earlier this month, the NFL struck a “landmark” deal with ESPN, trading control of the league’s broadcast properties for a 10 percent stake in the network. The agreement is most notable because of ESPN’s nearly-billion-dollar sports betting arm, ESPN Bet. The deal thus creates the closest entanglement yet between a league and a newer national pastime: sports gambling, an industry worth nearly $14 billion and commanding the attention of nearly half of men ages 18 to 49.

Yet as the two companies cozy up, Americans are growing more concerned about what sports gambling is doing to their country. In one recent poll, 58 percent wanted the federal government to “aggressively” regulate betting. Nearly two-thirds backed the SAFE Bet Act, a proposal from Democrats Rep. Paul Tonko and Sen. Richard Blumenthal that would impose stringent federal oversight. While a majority still favor legal sports betting, they also say that it should be illegal to bet on college sports, according to a recent AP-NORC poll.

This concern makes sense given the damage that gambling has done to both sports and bettors. Americans are recognizing the cascading harms legalization brought. Now they’re asking: Can we do anything to fix this mess? The answer is almost certainly yes. But it will require not only smart regulation, but the political will to overcome deeply entrenched financial interests.

Since 2018, when gambling became legal nationwide, professional sports have been beset by scandal. Most recently, two players on the Cleveland Guardians were suspended amid an investigation into whether they adjusted their play to make money on “prop” bets that they would behave in certain ways. But that’s just the latest: An ESPN timeline identifies dozens of incidents across college and professional sports in the past seven years, some involving millions of dollars in fraud.

That’s not surprising. Wherever players have the opportunity to make money through gambling, some of them will take advantage. Legal, online sportsbooks, with their deep markets and increasingly exotic bets, make doing so easier than ever before. And financial ties between the leagues and sportsbooks mean both parties are strongly incentivized to look the other way.

The rising tide of addiction and harm is less visible than the scandals, but far more harmful. An estimated 2.5 million Americans suffer from severe gambling addiction each year; 5 million to 8 million more have a mild to moderate gambling problem.

A growing body of rigorous research shows that legalizing betting systematically worsens the harms of addiction. Irresponsible gambling nearly quadruples following legalization, one June study found, followed by a 75 percent increase in calls to gambling helplines and increases in alcohol consumption. Legal sports betting increases credit card debt and risk of overdraft, another study found, and it increases the risk of bankruptcy by 25 to 30 percent, a third concluded. Gambling has also been connected to anxiety, depression, suicidality, and even domestic violence.

Many Americans now see a friend or loved one reflected in these statistics. In a recent poll, 18 percent of Americans reported that sports gambling has caused tension in their relationships; almost a third know someone whose well-being has been harmed by gambling. Since I first began writing about this issue, I’ve heard too many stories about people, mostly young men, tens of thousands of dollars in debt on betting apps. Such men are often those least able to afford a big bet; some will spend years or decades digging themselves out.

Such experiences are why Americans are clamoring for action. There are a lot of ideas out there for what that might look like.

Take the proposals in Tonko and Blumenthal’s bill: Curtail sportsbook ads, especially during live events; cap deposit frequency and prohibit credit card-backed deposits; prohibit the use of AI to track users’ habits. To those we might add bans on the most outlandish practices, like the discounts some sportsbooks offer for many-leg parlays. Or consider popular ideas like banning networks from broadcasting odds during games.

Lawmakers could also impose stiff fines on sportsbooks when players use them to profit off the game, creating an incentive for the operators to care about game integrity. Even if the feds don’t pick up these ideas, state legislators certainly can.

Regulation, though, is easier said than done. Because sports gambling is addictive, its harms concentrate among the small fraction of users who use most heavily. But these users also drive the large majority of sportsbooks’ revenue. Consequently, any regulation that gets at the fundamental problem will also be fiercely resisted by the industry.

Such problems are familiar to critics of industrialized vice, and overcoming them involves giving the public a clear harm done by a clear bad actor. Take the example of Candace “Candy” Lightner, whose 13-year-old daughter Cari was hit and killed by a drunk driver. Lightner’s organization, Mothers Against Drunk Driving, successfully passed hundreds of drunk driving laws, partly by telling the story of Cari and girls like her. Or consider America’s remarkable success at reducing smoking—thanks in large part to the villainization of Big Tobacco for its dissembling about cigarettes’ carcinogenic effects.

When people ask me if we will ever rein in sports gambling, I always give the same answer: once things get bad enough. From cigarettes to crack cocaine, our history with vices shows that we only start to take it seriously after the big tragedy happens. But as negative reports pile up, it seems like that story may be coming for sports gambling. Lawmakers—and the concerned public—need to be ready to act when it does, or expect more of the same.

Moynihan argued, I think correctly, that something like this had happened in America over the course of the 1960s, ‘70s, and ‘80s. A surge of crime, vagrancy, and family breakdown had exceeded society’s capacity to manage it, so instead we simply insisted that the problems weren’t real. He highlights, for example, declining concern with the prevalence of murder, unwillingness to recognize the behavior of the publicly seriously mentally ill as harmful, and the attempt to normalize non-traditional family structure. You may, of course, consider the normalization of these behaviors to be a good thing, but you should be willing to see the accuracy of the description nonetheless.

The idea of “defining deviancy down” gives us a useful way to frame any decay in norms. When we begin to accept previously deviant behavior, Moynihan suggests, it usually reflects some reduction in our capacity to manage that deviance. The normalization of deviance, in other words, is a policy choice. When it becomes more normal, we should look to see what institutional or structural capacity has eroded.

Subscribe now

A few weeks ago, my MI colleague Jesse Arm had a thoughtful piece at City Journal on the rise of what he termed “Luigism.” The piece was linked to the murder of Wesley LePatner, the Blackstone executive gunned down in the 345 Park avenue shooting. LePatner was probably not targeted for her professional affiliation, but that didn’t stop a horde of unhinged individuals from celebrating her death on social media, and calling for more of the same.

Many drew a connection between LePatner’s death and the murder of UnitedHealthcare CEO Brian Thompson. Thompson’s killer, Luigi Mangione, shot Thompson in cold blood in the middle of a busy Manhattan street because of the poorly specified evils of American health insurance. For this, he’s become a folk hero, with anonymous posters extolling his actions and often endorsing similar gestures.

Arm, and others, have identified this as an ideology—thus the “ism” in “Luigism.” But it’s not obvious that the ideology—a sort of antisocial anti-capitalism—is the most salient characteristic of these utterances. Merely disliking the rich should be distinguished from vocal support for their murder.

Perhaps it is better to understand Luigism as a kind of deviancy. Shorn of the justifications its proponents offer, Luigism is simply the affirmation of violence as a legitimate political means, and therefore a deviation from the implicit premise of democracy: that our problems are resolved at the ballot box, not with the bullet. We can ask how people justify this deviancy, but doing so fails to acknowledge the nature of the thing. I can tell you about the long history of leftist support for terrorism, but it would be a causal fallacy to do so.

Why not take the Moynihan view of things, then, and ask what the machinery that once managed the deviancy was, and when and how it was stopped? Why has the supply of bloodthirsty invective exceeded our supply of stocks and whipping posts, such that people are now comfortable wearing t-shirts emblazoned with the face of a murderer?

One obvious, unsatisfying answer is the internet. We generate more speech (written and spoken) today, for a larger audience, than at any point in human history. One of the unfortunate side effects of this is that people who hold fringe views will be able to much more efficiently search for and interact with other people who hold those views. This is how you get adult baby/diaper lover communities,1 and also how you get people who build cults around Luigi Mangione.

This answer isn’t just unsatisfying because it suggests the source of the problem is the source of every other social phenomenon. It’s unsatisfying because the internet has, over the past two decades, become easier, not harder, to regulate. As the average person has become more likely to be online, most social interactions have shifted from poorly managed fora to large social media sites, which are relentlessly moderated. That moderation is only made easier by the advent of large language models which, if nothing else, are extremely good at identifying the potentially problematic meanings of text or images.

So why don’t we moderate out this kind of content? The answer, I suspect your lizard brain is telling you, is “free speech.” Part of the ethos that Americans have brought to the internet is the idea that speech is good, more speech is better, and we should err on the side of permitting speech. So even when we have the tools to limit it, we should prefer free and open debate.

Yet this notion sits uncomfortably alongside the celebration and endorsement of criminal acts, especially criminal acts that are mala in se. Murder is illegal because it is wrong, and very few people will say that they think this should not be the case. Yet we tolerate people expressing their support for murder in often explicit and profane ways. Why? How is it that we came to think that “free speech” involves the protection of endorsement of murder?

The answer to this question actually has a somewhat important, and strange, history.

Subscribe now

TCF readers will recall that I wrote a few weeks ago about the hidden history of America’s hate speech laws. Most speech in the United States is protected by the First Amendment, meaning that regulations of it can only survive legal review if they meet the onerously high “strict scrutiny” standard. However, certain classes of speech—“low-value speech”—do not receive this protection. These include obscenity, “fighting words,” and defamation (which covers libel and slander). It turns out that the Supreme Court has blessed the idea libeling a group is “low-value speech,” and bans on group libel are basically hate speech laws if you squint long enough.

Another category of “low-value speech” is incitement—speech that provokes unlawful behavior or encourages people to act unlawfully. In the early 20th century, the Court was forced to answer a series of questions about when incitement—advocacy for anarchism, rebellion, and endorsement of other criminal acts—could be met with government suppression. Usually, the Court affirmed that this sort of speech was not protected. They usually applied what is called the “bad tendency” test, which held that “when the facts of a case indicate that the communicator intended a result that the state has prohibited, the court may reasonably assume that the communication has a tendency to produce that result. Furthermore, on the basis of that tendency, the court may punish the communicator for violation of the law.”

This test, founded in the common law the United States inherited from England, gave wide latitude to the suppression of dissident speech in cases like Abrams, Gitlow, and Whitney. These cases usually pertained to communist or anarchist speech, but they touched on the broader category of incitement.

Incitement is still “low-value speech” today. But since the “bad tendency” era, the Court has so substantially narrowed the definition of incitement as to render it impotent. In 1919’s Schenck, Oliver Wendell Holmes (writing for the majority) propounded the “clear and present danger” test, which established that inciting speech could only be criminalized based on its “proximity and degree.” The “clear and present danger” test never quite displaced the “bad tendency” test,2 but it opened the door to the idea that not all calls for criminal acts were unprotected.

Fast forward to 1969’s Brandenburg v. Ohio. The case, which concerned the conviction of a Ku Klux Klan leader under Ohio’s criminal syndicalism statute, became an opportunity for the progressive Warren Court to realize the free speech maximalist dreams of Holmes et al. In a per curiam opinion, the Court wiped clean the slate of incitement law, replacing all prior standards with the single, still regnant, “imminent lawless action” standard. Under Brandenburg, inciting speech could only be criminalized if it was “directed to inciting or producing imminent lawless action and is likely to incite or produce such action.”

The Warren Court’s motivations were in part a product of the moment. As Florida State’s Alexander Tsesis notes, the Brandenburg test “reflected societal consensus about the censorial overreach of congressional hearings and prosecutions against dissenters that occurred during the Red Scare period.” It is hard, moreover, not to see a connection between Brandenburg and the Warren Court’s apparent sympathy for the then-growing anti-Vietnam protest movement, which might otherwise have been quashed by the same laws that were deployed by the Wilson administration during WWI and upheld by the Court at that time.

After all, this was the same Court that enshrined the right to wear the words “fuck the draft” on a leather jacket in public, because, as the majority put it, “one man's vulgarity is another's lyric.” One might see an analogy to the case of Brandenburg and consider that one man’s promise to take “revengeance” against “black peoples” and “Jews”—that being what Clarence Brandenburg was arrested for—is another man’s legitimate political discourse, apparently worthy of First Amendment protection.

Subscribe now

Why, then, do we think it is vital to allow people to celebrate murders on social media? At least some of that speech could be regarded as having a “bad tendency”—they tend to incite more murders. Yet the narrowing of the domain of incitement has moved the vast majority of such speech under the First Amendment’s protection. More to the original point, since 1969 we have taken away the power of our institutions to regulate such speech. Moynihan’s theory tells us that the next step is for us to stop defining that speech as deviant. And just as Moynihan’s theory predicts, we have done so.

The thing that alarms the average person about “Luigism,” more than its often amorphous ideological predicates, is that it is (at least sometimes) incitement. The underlying arguments are perverse. But the underlying arguments can be advanced by people who do not end up endorsing murder. What differentiates those celebrating the murder of Wesley LePatner is that they are celebrating murder—the sort of deviant behavior that we have rendered ourselves powerless to stop.

Free speech maximalists might argue that Luigism is the price we pay for being allowed to express other dissenting views freely.3 To put it in drier terms, minimizing false positives will necessarily net you a lot of false negatives. In the domain of speech, we tend to assume that false positives are a lot worse than false negatives, insofar as speech is a right, and the deprivation of rights is a greater harm than the harms done in their exercise.

Part of the point of “low-value speech” as a category, though, is that certain kinds of speech do not serve the ends free speech is meant to serve, which is to say open political debate in a free republic. Incitement is not and has never been protected speech, and for most of our nation’s history we lived just fine with a definition of incitement that would include some of what Mangione’s supporters say. I understand that some take the view that celebrating a woman’s murder is part of open political debate. I do not agree.

A different way to frame the maximalist argument is that the extremity of speech is in the eye of the beholder. Prohibiting endorsement of murder as incitement is, in this view, a slippery slope to the regulation of all sorts of political speech. The view of the modern right-wing free speech advocate—embodied, for example, in J.D. Vance’s criticisms of the European approach to speech—is that, insofar as liberal institutions dominate our society, they will tend to proscribe acceptable right-wing views.

As someone who holds such views, I find this prudential case for free speech appealing. At the same time, as someone who holds the (right-wing?) views that words have meanings and truth can be discerned through the application of human reason, I chafe at the kind of relativism that this argument presumes. The problem with the idea that every right-wing utterance is “harmful” or “violence” is not that words can’t be harmful—they can. It’s that most of those utterances do not rise to that level. And, more to the point, it is hard for me to accept relativism when it comes to judging whether or not it is wrong to endorse the brutal murder of innocents.

Moreover, the celebration of LePatner’s death is far from the only case where speech that endorses crime is a pressing matter of public debate. The rise of the internet has, for example, created the problem of “terroristic incitement”—people using the web to encourage other people to commit terroristic acts. This behavior probably can’t be criminalized, even as it creates a grave public security threat. Similarly, there is a live debate about whether and under what circumstances non-citizens have which free speech rights—a debate which, most honest brokers admit, is not settled.

Subscribe now

Regardless, I think the Court is unlikely to revisit Brandenburg any time soon. Perhaps the greatest critic of the maximalist position on the Court, Justice Alito, appears to have capitulated to the view in at least some cases. And the center of the Court’s general predilection for stare decisis gives little reason to expect any radical change.

Still, there are at least two arguments that Brandenburg need not be the final say on the matter. One is the straightforward Originalist argument. At the Founding, many probably took “freedom of speech” to mean what Blackstone thought it meant, i.e. a prohibition on “prior restraint” or censorship. Justice Joseph Story, the 19th century Supreme Court Judge whose Constitutional interpretation is often taken as definitive, endorsed this view, with the recognition that government could punish speech that “injure[s] any other person in his rights, person, property, or reputation” or “disturb[s] the public peace, or attempt[s] to subvert the government. The “no prior restraint” interpretation was the view the Court advanced in many of its early cases, and indeed as late as 1907.

The notion of “low-value speech” was an adaptation to the death of this earlier view, itself the result of innovation in the Court in the middle of the 20th century. Even if one takes a more expansive view of the First Amendment, it is really quite hard—as Tsesis argues—to support Brandenburg specifically based on original public meaning Originalism.

The second argument is that, much as in the case of obscenity, the playing field has changed. The internet really has loosed a great deal more deviant speech than the Court could possibly have contemplated in 1969; after all, who wanted to listen to Clarence Brandenburg, anyway?

Yet if I could have told Earl Warren that his ruling would help advance a world in which thousands of people would publicly celebrate the murder of an innocent woman, would he have decided the same way? I’m not so sure. And a revised standard need not look like the “bad tendency” test in order to have more teeth—to acknowledge, at least in principle, that support for the bloody murder of innocents is not speech worth strict-scrutiny protection.

If you do not find this normative case compelling, that’s fine. We should still agree on the descriptive. This sort of deviance has become more common because we have both increased its supply and, simultaneously opted to reduce our power to reign it in.

But I don’t actually know what you read me for, or why you read me. I also don’t know what sort of content you want from TCF. But I am a loyal servant of the content gods, so I am posting this series of polls to try to get some ideas.

If you like TCF, please take two minutes to answer the below!

I would appreciate additional comments in the replies. Vote early, vote often!

Pirate Wires has also been kind enough to republish some of my work on disorder, as well as originally running the piece below, on the coming crisis in New York City incarceration. This is a little shorter than my usual offerings on Substack, but I’m reposting it here at their request, to give you a taste of what they’re is about. If you’re interested, check them out too.

Zohran Mamdani — the 33-year-old former amateur rapper, Democratic Socialist, and likely next mayor of New York City — has a lot of ideas. But the most impactful thing Mamdani might do as mayor isn’t freezing the rent or making buses free (it’s not clear the mayor can even do that stuff). Rather, Mamdani could enter Gracie Mansion and enact a long-held socialist dream: shuttering New York City’s jail system.

Under current law, New York will close the infamous Rikers Island jail complex in August of 2027. The facilities meant to replace it, everyone agrees, are unlikely to be ready by then. Yet somehow, no one seems interested in addressing the impending crisis — especially not Mamdani, a committed prison abolitionist. In fact, to free 7,669 people currently held before trial — mostly charged with crimes like murder, robbery, and rape — all Mamdani will have to do is… nothing.

How is this possible? Roll the clock back to 2019, when the New York City Council passed a law requiring that Rikers close. In its place, the plan is to build four jails in the city proper, one in each borough except Staten Island. That plan reflected the idea that Rikers was dangerous, dysfunctional, and cruel (fair, given that it’s now in federal receivership). It was also a triumph for the then-ascendant criminal justice reform movement.

But while Rikers has problems, the idea of building four massive jails on some of the most expensive real estate in the world in just eight years was always implausible. It only became more unrealistic as the Covid pandemic and Biden inflation drove up labor costs and wait times. Earlier this year, the city commission that first called for closing Rikers admitted there's no way the city will hit its 2027 deadline. The Brooklyn jail is now expected to open in 2029, the Bronx jail in 2031, and Queens and Manhattan’s facilities in 2032.

In other words, New York City legally must close Rikers in August 2027 with no suitable replacements. Mayor Eric Adams has agitated for extending the timeline, but found himself at an impasse with the City Council; he’s now floating repairing, instead of closing, Rikers.

Into this mess, add Zohran Mamdani.

Mamdani, among his other radical bona fides, is a committed prison abolitionist. He’s repeatedly called to “abolish” incarceration, writing on X that “the entire carceral system is an unreformable public health hazard.” Asked whether prisons are “obsolete,” Mamdani replied in an unearthed clip, “what purpose do they serve?” He’s also called violence “an artificial construction.”

Mamdani has walked back some of his prior radicalism on the campaign trail. But he told The City that he supports closing Rikers, and would work with DAs to “release more people pretrial or divert them from prosecution entirely.” That’s typical of a committed criminal justice system reformer who, for example, once voted against punishing people who assault transit workers.

Years of reform mean that almost everyone who can be diverted from Rikers already is. As of this writing, there are only about 6,300 people detained pre-trial in NYC jails on Rikers (another 1,400 or so are being held there serving short post-conviction sentences, or they’re parole violators, or in transit to go upstate). 1,550 of them are in on murder or attempted murder; another 3,200 are in on robbery, assault, burglary, firearm or sex offenses. Under the state’s infamously lenient bail laws, 100 percent of Rikers detainees have been ruled a flight risk. Does Mamdani plan to put these people back on the street?

Some may expect a future Mayor Mamdani to accelerate construction of the four replacement jails, but that’s unlikely. The city’s recently passed budget deprives it of funds it could use to hasten the process. And even if it had the money, New York is three years past deadline and $100 million over budget on another 104-bed jail inside of Bellevue Hospital. If New York can’t even open a 104-bed facility, what hope is there it can expedite four new buildings with 30 times the capacity?

Where will all these criminals go, if not Rikers? Legally, New York City can ship its detainees to other counties’ jails. About 1 percent of prisoners statewide were originally held in NYC before being transferred. In theory, Mamdani could try to shift thousands of detainees to Westchester or Long Island — but doing so would be hugely expensive (because the city would have to pay other counties for their beds) and unprecedented in its scale.

It’s easier to imagine Mamdani standing on principle, and attempting to divert everyone flowing into Rikers, instead releasing them on their own recognizance or with minimal check-ins and other monitoring. Doubtless, the resultant crime spike will make him and the Council change course. But we’ve been given no reason to believe the crisis will be averted before it happens.

No matter who wins the mayoral race, the Rikers time bomb will keep ticking. New Yorkers need to ask mayoral candidates how they plan to defuse it. By all accounts, the leading candidate seems ready to let it explode.

The opening substantive post, from staff writer (whose departure from Vox means I no longer have someone to identify as the reasonable person at Vox), concerns the recent, consistent failure of guaranteed income pilots to produce any non-pecuniary benefits for their treated participants. This finding is, as Piper puts it, “shocking”:

Homeless people, new mothers and low-income Americans all over the country received thousands of dollars. And it's practically invisible in the data. On so many important metrics, these people are statistically indistinguishable from those who did not receive this aid.

I cannot stress how shocking I find this and I want to be clear that this is not “we got some weak counterevidence.” These are careful, well-conducted studies. They are large enough to rule out even small positive effects and they are all very similar. This is an amount of evidence that in almost any other context we’d consider definitive.

Piper is nonetheless undeterred in her optimism about more targeted transfers, and about social policy more generally. This optimism—the more precise term might be meliorism—is characteristic of the liberalism that Piper, Demsas, et al. claim they would like to revive.

Yet it is hardly unprecedented for that optimism to run aground on the realities of policy evaluation. Indeed, liberal meliorism has faced precisely the same challenge before, decades ago. And the same basic explanations for why still obtain.2

Subscribe now

Many people who have problems—emotional, psychological, interpersonal, etc.—also do not have money. One reasonable inference is that the cause of those problems is the lack of money. A solution to many of our social problems, therefore, might be to do as Annie Lowrey once enjoined, and “give people money.”

But what actually happens when you give people money? A series of increasingly large, increasingly well-funded randomized controlled trials have in recent years attempted to answer this question. These target slightly different populations, but they all work essentially the same way: recruit a large random sample; split them into treatment and control; give the treatment group a large amount of money per month (between $333 and $1,000) and give the control no or a token amount of money per month; see how the two groups compare.

What do these studies show? As Piper summarizes it, “guaranteed income transfers do not appear to produce sustained improvements in mental health, stress levels, physical health, child development outcomes or employment. Treated participants do work a little less, but shockingly, this doesn’t correspond with either lower stress levels or higher overall reported life satisfaction.”

Just to elaborate: the primary effect of transfers is that recipients have more money than non-recipients. Secondarily, recipients work slightly less, because they replace some but not all of the lost income with program dollars. But … that’s about it. Neither participants’ physical nor mental health improves. They don’t get more education or build skills. Their broader financial position doesn’t improve. Their parenting doesn’t improve. Their political participation doesn’t improve.

Why is this? Piper suggests that perhaps the improvements just aren’t showing up in the measurements:

Maybe this isn’t as confusing as it initially feels. If you gave me an extra 10% of my income to spend, it would absolutely make me better off — just not necessarily in ways you could measure.

I might give some to friends going through a time of need, which makes my community stronger even if it doesn’t show up in a study looking at my habits. I might take the kids across the country to visit my aging grandparents — stressful, but worth it in a way that probably doesn’t show up on a mental health inventory.

This is, and I don’t know how else to put it, a silly argument. Null effects show up across more or less every non-pecuniary measure across multiple RCTs. Blaming measurement error is unfalsifiable. Your response shouldn’t be “no randomized controlled trial is really going to convince me that money does not improve people's lives.” When the evidence so overwhelmingly says that money doesn’t change non-pecuniary outcomes, you need to change your assumptions.

And this is particularly so because big, bold nulls are not actually that unusual in the world of policy evaluation! In my own primary area of focus, criminal justice policy, transfers sometimes reduce monetary crime (mostly either by substitution or by getting people off of drugs), but basically don’t affect violent crime. Educational interventions? Often nulls. Or take something as simple as giving people health insurance: the famous Oregon Medicaid experiment found “no significant improvements in measured physical health outcomes” in the treatment versus control groups.

The prevalence of nulls in evaluations of social policy interventions is so well-established that there’s actually a term for it. It’s called Rossi’s Iron Law, after sociologist and policy evaluator Peter Rossi. Rossi’s iron law is: “the expected value of any net impact assessment of any large scale social program is zero.” There are a handful of corollaries, like the stainless steel law (“the better designed the impact assessment of a social program, the more likely is the resulting estimate of net impact to be zero”) and the brass law of evaluation (“The more social programs are designed to change individuals, the more likely the net impact of the program will be zero”).

Subscribe now

Rossi is a funny figure, actually. In addition to his work in policy evaluation, he was an ardent liberal of the Piper variety. His expertise in evaluation came partly out of his work on War on Poverty programs, part of the broader Johnson-era, Great Society push for government to do more to ameliorate the condition of the poorest Americans. His association with policy pessimism, in other words, is in spite of his politics and his vocation.

Yet Rossi’s formal pessimism was not atypical for the liberal establishment of his era. The Great Society was meant to work the kind of transformation that UBI advocates sometimes allude to. The goal, importantly, was not just to give people money, but to help people—through transfers and also government programming—overcome the misery and dysfunction which alarmed leaders like Johnson. They wanted to make “taxpayers out of taxeaters,” in an evocative Johnsonian turn of phrase—to help citizens become independent through government support.

At the macro level, more or less the opposite obtained. Beginning in the late 1960s, social pathology exploded: crime, drug use, family breakdown, teenage pregnancy, urban collapse, and a general personal and social apocalypse. Concurrently, many social scientists observed that many social policy interventions did not obtain their desired goals. Charles Murray, the social scientist and prominent public controversialist, made his name arguing that the Great Society had caused all of these dysfunctions, by subsidizing the lifestyle of the lower classes. On the back of this theory was borne welfare reform, among other initiatives.

Personally, I am less partial to the Murray theory and more to one that might be identified with Rossi, but more likely with the public intellectual Senator Daniel Patrick Moynihan. In the report that launched him to infamy, Moynihan—then a deputy at the Johnson labor department—argued that the problems then faced by black America were the result of a “tangle of pathology,” itself downstream of a deep sickness of family structure with roots in slavery. Moynihan’s slavery argument was probably wrong (William Julius Wilson, another noted progressive sociologist, demonstrated this later). This is especially so because the tangle of pathology has grown to encompass much of white America, a phenomenon documented both by Murray and the Vice President.

Which is not to say that guaranteed income experiments fail because of family structure per se. Rather, the point is that the implicit causal model of these experiments (less money → more problems) is almost certainly wrong. One or more additional, hard-to-alter variables is or are likely causing both people’s lack of money and their other problems, such that addressing one does not address the others.

What is that variable? There are lots of explanations. If you’re Murray you talk about biology; if you’re Moynihan, you talk about family structure; if you’re William Julius Wilson, you talk about deindustrialization. Heck, there are even left-wing variants here: “critical race theory,” before it was an object of cultural discourse, referred to the left-wing legal tradition that was argued that the civil rights revolution had not ameliorated the black condition in America, which was taken as foundational evidence that deeper, more “structural” racism may be at play.

Which of these explanations is correct? Well, that’s … a causal fallacy. I tend to prefer a simpler formulation: for whatever reason, it is very, very hard to change people using policy. (This is Rossi’s brass law.) If you give people money, their bank accounts will be a little more full, and they might work a little less to compensate. But they won’t become healthier, or better educated, or more responsible with money, or better parents, or better citizens. They’ll be the same people with a little more cash.

This doesn’t mean that policy can’t do things. It can, and often does. But as a general rule, when it does it does so through simple channels that involve shifting people’s incentives, rather than trying to change people as people. And because those channels are few and far between, we should expect policy that tries to do the hard work of changing people to mostly fail.

Subscribe now

There’s actually a term for this line of critique, as applied not merely as an evaluative insight but a political one. That term is “neoconservatism.” Long before it took on today’s pejorative sense,3 it referred to a group of previously left-of-center intellectuals who were persuaded, largely by their encounters with actually existing government policy as implemented under the Great Society, that policy rarely worked, often backfired, and reliably could not change who people are.

This original group of neoconservatives—Moynihan, James Q. Wilson, Irving Kristol, Ed Banfield, and so on—more or less decisively won the policy debate in the 1980s and 1990s, such that liberals like Rossi (and William Julius Wilson, and James Coleman, and Christopher Jencks) were often forced to grant their premises. Indeed, part of how we ended up at the idea of unconditional cash transfers is that giving people money became something of an alternative to the more comprehensive social services the Great Society had once envisioned, but which fell out of fashion when it seemed like they weren’t working.

All of this intellectual history brings me back, in a roundabout way, to Piper and her shock at the failure of guaranteed income. If you are, as a matter of principle, committed to the idea that giving people money will solve their problems, then I cannot help you. But if you are willing to see the evidence that Rossi was right, and that social policy can rarely change people, then the failure of UBI should be for you what Irving Kristol famously referred to as the experience of being “mugged by reality.”

You can, that is, insist that the evidence has failed your principles. But in my experience, if the evidence is against your principles, it’s not the evidence that may need to change.

You may like or dislike these laws (that’s not really the subject of this post), but it is hard to argue that they don’t serve their intended purpose. A 2018 paper, Cunningham et al., looked at the effect of a TRAP law imposed in Texas in 2013. Its authors argue that the resultant clinic closures and increase in average distance to a clinic reduced abortion rates by 15 to 40 percent, depending on clinic distance. The above chart is from another paper, Arnold 2022, which looks at the average effects of the first TRAP law implemented in a state using a difference-in-difference specification. It finds the laws yield “a reduction in the abortion rate of approximately 5% the year the first law is implemented, and an average reduction of 11-14% in subsequent years.”

That TRAP laws reduce abortion should be unsurprising to TCF regulars. As you may remember, prohibition is a form of regulation. The obverse is true: regulation is a series of narrowly tailored prohibitions. As I wrote previously:

To reiterate, the more general form of this argument is that prohibition is a form of regulation. They are not distinct kinds of thing; one is a particularly aggressive instance of the other. Indeed, many kinds of regulation are simply targeted prohibitions. A hypothetical national potency cap is a prohibition on the sale of marijuana which exceeds the cap. A labeling requirement is a prohibition on the sale of marijuana which fails to comply with the requirement. Taxes on potency are a prohibition on the sale of untaxed product. The fact that some prohibitions are more or less targeted does not make them meaningfully distinct.

The same logic holds for abortion. An abortion ban (as exists in 12 states) is a prohibition on abortion. Like prohibition of other commercial services, it doesn’t reduce the level of abortion to 0, although it plausibly reduces it. Similarly, regulations of abortion are prohibitions on abortions that violate those regulations. TRAP laws prohibit non-compliant abortion providers, and therefore prohibit abortions provided by those providers. As a result, they reduce the total level of abortion in the equilibrium. I will reiterate (because this is a contentious topic) that one does not need to believe that this is good in order to believe that it is happening.1

If you are the sort of person who opposes abortion, though, then the utility of TRAP laws is that they shift the discourse. You no longer have to have a debate about the merits of prohibiting abortion (which, as of 2025, you are probably going to lose). Instead, you get to talk about health and safety standards in abortion facilities. You can point out that some facilities are probably below standards that sound reasonable (why shouldn’t abortion providers need to have admitting privileges at nearby hospitals?), and you can use graphic stories about particularly poorly administered Planned Parenthoods (or whatever) to make your point. Moreover, because you have shifted the discourse to topics of health and safety, a large fraction of the public will simply tune out. Abortion is a hot-button social issue. Surgical facility standards are not.

With all this in mind, let’s talk about pornography.

Subscribe now

As I observed over at The Dispatch recently, a great deal of pornography distribution is probably already illegal. That is to say, some substantial fraction of porn available online right now probably meets the Supreme Court’s extremely high Miller standard for “obscene” content. Obscene content is unprotected by the First Amendment and illegal to manufacture, distribute, or otherwise profit from under 18 USC Ch. 71 and a whole host of state laws. (Section 230 also doesn’t offer protection for platforms.) Consequently, I argue, the state has all the power it needs to, for example, shut down Pornhub.

I think the reason policymakers won’t do this is the polling—most Americans are opposed to banning pornography. That was true when I did this roundup of data on porn regulation for IFS six years ago (crazy how the time flies). And it remains true today. In the 2024 General Social Survey, fewer than 30 percent of respondents supported an all-out ban, compared to over two thirds who want it banned just for minors. We don’t like to say porn is morally acceptable, and we sure don’t like to admit to using it. But we also don’t want it banned.

But, despite my protestations, we regard regulation and prohibition as distinct. And Americans do want pornography regulated. That was true in my IFS review. And it remains true in the polling from AEI/the Survey Center on American Life above, which finds that “[n]early seven in 10 (69 percent) Americans support making internet pornography less accessible, while fewer than one in three oppose such efforts.”

As with abortion, regulation of porn shifts the discourse. After all, banning sounds lame and square; regulation sounds reasonable and necessary. That’s particularly true of an industry that keeps facilitating the distribution of child sex abuse material, rape content, and revenge porn.2 How can people oppose reasonable regulations of the porn industry?

But of course, regulation is a form of prohibition. Imagine TRAP laws for porn—TRPP laws. These would be targeted regulations of pornographic content providers which are facially reasonable, but which impose substantial compliance costs that result in a reduction in availability of the product.

Actually, you don’t have to imagine. You can just look at North Carolina, where they’re currently trying to pass HB805 over an executive veto.3

HB805 does a number of things, including a bunch of gender-identity-related actions (these are probably why Governor Stein, a Democrat, vetoed). But in relevant part, it requires that online operators that publish or allow users to publish pornographic material verify that everyone appearing in that material “has provided explicit written consent for each act of sexual activity in which the individual engaged during the creation of the pornographic image”4 and “has provided explicit written consent for the distribution of the specific pornographic image.” It also includes rules about the consent forms, including requiring the presentation of a driver’s license or other identification. Failure to comply can result in a $10,000 fine per day of non-compliance; depicted people who did not provide consent can also bring civil suits.

I mean… this is a TRPP law. There are facially good reasons for these rules—it really is a huge problem if pornographic images of people are being distributed without their consent. And it’s reasonable to demand that distributors make sure that this isn’t happening. But also, there are billions and billions of porn “images” available online, many with totally anonymous participants. The compliance costs for a distributor like Pornhub are essentially impossible to meet. Pass the TRPP law, and the supply of porn will go down.

The big open question here is not popular support — I think you could impose similar laws in other red states — but constitutionality. Non-obscene pornography5 is protected speech, and the Court has repeatedly looked askance at the burdening of adult access to protected speech for other purposes. In legal terms, attempts to do so must survive “strict scrutiny,” the highest standard of Constitutional review, which says speech rights can be burdened only if the law is “narrowly tailored” to further a “compelling government interest” and uses the “least restrictive means” to achieve that interest. There are only two cases where the Court has blessed a restriction on speech under this standard (although they are notably both in the past 15 years.)

But there are signs that the Court might be shifting its posture on porn. In the recent ruling on age verification laws (Paxton v. Free Speech Coalition), the majority defined the problem in such a way as to be able to subject Texas’s verification law to a lower standard of review (intermediate scrutiny) which it successfully passed. Justice Elena Kagan, speaking for the three liberal justices, insisted that the Court should have used strict scrutiny … but implied that the laws might have passed muster anyway. That’s consistent with the view, going all the way back to the dissents in 2002’s Ashcroft, that the speech law made for obscenity in the 1970s might not be good for the technology of the then-future, now-present.

There is an analogy here, too, to TRAP laws. In 2016’s Whole Women’s Health v. Hellerstedt, a relatively liberal Court struck down a Texas TRAP law. A similar result obtained four years later, in June Medical Services, LLC v. Russo. In that case, though, it was only because of Justice Roberts’s commitment to stare decisis, or deference to precedent. Roberts joined in the ruling with the four liberals, but reiterated his dissent from the decision in WWH. The other four conservative justices on the Court dissented in full. With the Court’s new 6-3 majority, one suspects that these precedents might be in danger of falling, too.

As with abortion restrictions, so with pornography restrictions—a new Court means a new political calculus. Claims of the “substantive due process death of obscenity law,” in other words, may have been premature. And now is the time for states to begin asking the Court tough questions about exactly how far they can go on the regulation of porn.

After all, unlike TRAP laws (which are divisive), the American public would like to see regulations of pornography. They are unlikely to be against consent requirements of the sort North Carolina is likely to impose. All that’s left, if the state actually passes its law, is for the Court to bless it; I am not certain, but I think there’s a real chance they will.

While it is commonly accepted, it is also not precisely true. While many of the hate speech prohibitions in other nations would not fly here, what are known as “group libel laws”—which prohibit the defamation of groups of people based on their inherent characteristics—have the blessing of the Supreme Court. Such laws remain on the books, even if never enforced, in several states.

Indeed, there was a brief, now mostly forgotten, period in American life when group libel was expected to be the front-line defense against rising fascism. And while the laws have faded from public discussion today, they raise important questions about how we have come to think about speech in the modern era, and the role of speech in public life.

Let’s back up and talk for a minute about defamation. “Defamation,” the Free Speech Center’s First Amendment Encyclopedia helpfully informs us, “is a tort that encompasses false statements of fact that harm another’s reputation.” There are two kinds—spoken defamation (slander) and written defamation (libel). Both types secure, as Justice Potter Stewart put it, “the right to protect one’s good name.” The law also distinguishes civil libel—a civil claim brought by one party against another—from criminal libel, which is enforced by the state, and justified by libel’s tendency “to create breaches of the peace when the defamed or his friends undertake to revenge themselves on the defamer.”

Historically, libel was not considered speech protected by the First Amendment. In Chaplinsky v. New Hampshire, defamation was listed along with obscenity, “the profane,” and fighting words as falling outside the bounds of Constitutional protection. Since 1964’s New York Times Co. v. Sullivan, the Supreme Court has placed First Amendment restraints on libel law, in particular making it harder for public figures to bring libel claims. But it remains possible to bring libel suits today, and criminal libel laws remain on the books in a number of states.

Libel is not restricted to individuals. Although the Supreme Court has imposed some constraints, traditionally one can libel a business, or the government (seditious libel) or God (blasphemous libel).1 This raises the question: can one libel a group of people?

In the 1940s, a major social movement emerged built around the idea that the answer was “yes” and, moreover, that “group libel” laws were an effective way to fight what we would today call hate speech.2 While civil group libel posed challenging legal problems—who has standing to sue on behalf of an entire group?—criminal group libel laws seemed to stand on firmer ground. Perhaps states could criminally punish people who incite hate in a way that creates breaches of the peace. In 1917, Illinois passed the first such law, prohibiting the distribution of any publication which “portrays depravity, criminality, unchastity, or lack of virtue of a class of citizens, of any race, color, creed or religion” and thereby “exposes the citizens of any race, color, creed or religion to contempt, derision, or obloquy or which is productive of breach of the peace or riots.”

The group libel movement did not really get going until the rise of fascism in Europe, when many Americans identified fascist ideas as a serious threat to our democratic way of life. As legal historian Samantha Barbas writes:

The ascendance of the Nazi regime in Germany in the 1930s and the proliferation of Nazi groups in the U.S. led to calls to criminalize antisemitic defamation. In the 1930s, hundreds of Nazi groups formed in the U.S., with names like the Silver Shirts, Defenders of the Christian Faith, and the German American Bund, with over 25,000 members. Many of these were aided directly by Hitler’s government. By 1939, there were 800 pro-fascist or pro-Nazi organizations in the United States. … One-third of Americans were said to receive fascist literature regularly in the mail in the early 1940s.

Many were alarmed by both the content of these messages and the idea that a hostile foreign power was attempting to influence domestic concerns. In response, prominent Americans argued that group libel would be an effective way to control such speeech.

It is hard to overstate the influence of this movement—supporters, Barbas writes, included “the American Jewish Congress, labor organizations, Communist Party members, liberal academics, and public officials, including Solicitor General Francis Biddle, who in a 1940 address before the American Association of Law Schools advocated group libel law.” By 1950, there were group libel laws in California, Connecticut, Illinois, Indiana, Massachusetts, Nevada, and West Virginia, as well as in the cities of Cincinnati, Chicago, Sacramento, Denver, Houston, Oklahoma City, Omaha, and Portland, Oregon.

That did not mean, of course, that there was not opposition to group libel laws. Much of the nation’s press opposed them, as did the ACLU. And while there were Jewish and black groups in favor of group libel, there were also Jewish and black groups which fervently rejected them. Opponents generally argued that group libel laws contravened the First Amendment, and that the correct way to fight bad speech was with more speech, not less.

The legal dispute came to a head in 1952’s Beauharnais v. Illinois. Joseph Beauharnais, president of a white supremacist group, had distributed a pamphlet which called for “one million self-respecting white people in Chicago to unite” against the supposedly malign influence of black people, adding that “if persuasion and the need to prevent the white race from becoming mongrelized by the negro will not unite us, then the aggressions … rapes, robberies, knives, guns and marijuana of the negro, surely will.” He was arrested, tried, and duly convicted under Illinois’s group libel statute. On appeal, Beauharnais argued that the law violated his First Amendment rights.

That argument was not remarkable; after all, the ACLU had been saying much the same for years. What was remarkable was the resultant ruling in Beauharnais. In a 5-4 decision, the Court concluded that Illinois’s group libel statute did not run afoul of the First Amendment; group libel, in other words, is not protected speech. The Court relied on its finding in Chaplinsky that libel was outside the First Amendment’s protection, adding that “if an utterance directed at an individual may be the object of criminal sanctions, we cannot deny to a State power to punish the same utterance directed at a defined group, unless we can say that this is a willful and purposeless restriction unrelated to the peace and well-being of the State.”

Surprisingly, Beauharnais turned out not to be a stepping stone for the group libel law movement, but its capstone. By the time the Court handed down its opinion, the movement had largely dissolved, with the struggle for civil rights directing its attention elsewhere. Illinois repealed its law in 1961. Today, group libel laws remain on the books in Connecticut, Massachusetts, and arguably Nevada, but I have been unable to find any evidence of their being enforced.

As a result, Beauharnais remains mostly a constitutional oddity, an exception to the general rule that hate speech is permissible. Its effects, moreover, have been blunted by subsequent rulings. Criminal libel laws, once the foundation of libel law, have fallen out of legal favor. In particular, any group libel prosecution today would probably need to comply with the Court’s finding in Garrison v. Louisiana that truth is an absolute defense against criminal libel prosecution—requiring, therefore, a public adjudication of the truth of bigotry.

Why relate this story? In part, as an interesting piece of historical trivia—a whole movement mostly forgotten, but for a mark on our constitutional landscape. (And I am very interested in forgotten movements.) And now you can rebut your friends who claim that America has no hate speech laws—we have *almost* no hate speech laws.

But do I want to mount a defense of group libel laws? I suspect that in practice they would be mostly a fig leaf, and apt to be abused by the sort liable to challenge anything they don’t like as hate. (Though this second concern is in some senses obviated by the first, since truth would be a defense against prosecution.)

At the same time, I think Beauharnais, and the theory of speech which undergirds it, say something important about how we think about the role of speech in public life. I think our rejection of that older way of thinking has led to much of our contemporary confusion around speech as such, and that recalling what we used to think speech was is instructive for contemporary debates.

Subscribe now

I first learned about group libel laws from political theorist Hadley Arkes’s excellent book The Philosopher in the City: the Moral Dimensions of Urban Politics. The book is about philosophy, law, and urban life (you can imagine why I was interested in it). The first two chapters are dedicated to group libel and the phenomenon of speech generally, and are an extension of prior writing Arkes had done, for example, objecting to the American Nazi Party being allowed to march in Skokie, Ill. This latter argument was first published in National Review,3 and Arkes is generally identified as a conservative. From the vantage point of 2025, this seems unusual—aren’t conservatives supposed to be all about free speech?4

Arkes was, however (and still is) defending an older view of speech against what he perceived as judicial assault. Following Beauharnais, in the period during which Arkes first started advancing his arguments, the Court oversaw a general legal and cultural transition to what my Manhattan Institute colleague Tal Fortgang has labeled “free speech supramaximalism,” the idea that “not only must speech prevail over regulation, but nearly everything is sooner or later described and defended as speech.” It took this position when asked about hate speech, which is generally protected. But it’s also took it on the variety of activities that are now demarcated as speech, from flag burning to wearing the words “fuck the draft” to letting Nazis march in neighborhoods full of Holocaust survivors.

Against this attitude, Arkes preferred what he regarded as the traditional understanding of speech, grounded both in English common law and what the Founders probably understood “freedom of speech” to entail. On group libel, and libel generally, Arkes writes:

The question … is whether injuries (or “injustices”) can be inflicted through the use of speech; and for most of our history the answer to that question seemed obvious. There was nothing about speech itself that rendered it categorically innocent or incapable of doing harm. … If there is an obligation then to restrict speech, even in a democratic regime, it is because speech can be a medium of injustice along with a variety of other devices. And if we have the capacity to judge whether harms are inflicted justly or unjustly, we do not lose that capacity to judge when the harms are inflicted through the use of speech.

Even as the Court has moved away from Arkes’s view, continuously expanding the zone of legally permissible speech, the idea that words can wound is in many senses stronger than ever. As Barbas, the historian, notes, concern with hate speech did not vanish with Beauharnais. It was revived by, among others, the legal scholar Richard Delgado, a godfather of Critical Race Theory who argued that hate speech should be addressed by the creation of a new tort action to permit those harmed by hateful words to recover damages.

Delgado’s successors—academic and activist alike—have sought to enshrine legal and social constraints on speech they deem to be harmful to an ever-expanding array of identities and classes. Influenced in part by them, today’s young people are famously censorious.

The result is something of a paradox. Legally, we have never been more permitted to speak. Socially, we are not at our most restricted, but we are almost certainly more restricted in our speech than the Court probably expected we would be when it started handing down liberalizing rulings in the 1960s and 1970s. This is how we get the routine spectacle of students exercising their free speech rights to protest in favor of silencing others. 50 years ago, we made sure that Americans had the right to wear a jacket that read “Fuck the Draft”; today, extreme social penalties accrue for all sorts of less obviously offensive sartorial choices.

Many, of course, see these two tendencies as being in tension. Yet lots of people seem to be able to hold both views simultaneously, which should cause us to ask what they have in common.

One answer is that they both justify themselves by reference to the individual. Advocates of free speech supramaximalism regard it as an offense against personal liberty to constrain anyone from saying anything at any point. The “woke,” by contrast, regard the individual’s right to be free from the barbs of hate as paramount—a principle which still focuses on the individual and his rights. As Barbas puts it, modern hate speech opponents “focused less on the social unrest caused by group defamation and more on the psychic and emotional harms that group defamation caused to individual members of minority groups.”

By contrast, neither side expresses objective principles for determining when speech is good. Neither side is really interested, that is, in any account of what the protection of free speech is for—which, if articulated, would permit us to say what speech is not for. One side argues that speech is good because it is speech, the other that speech is bad because it can harm. But when are speech’s harms justifiable? The law, historically, has considered this a cognizable question. But neither side has a coherent answer.

I might go so far as to suggest that our paradox—free speech maximalism alongside censoriousness—is a function of the lack of any sort of ordering principle or principles. Unlimited free speech will breed a backlash if we can’t make recourse to some kind of shared, coherent logic underwriting the problem cases that inevitably arise. Instead, we swing wildly between the two extremes because we have no method for calculating the golden mean.

Moreover, because the state has absolved itself of any role in the regulation of speech—even in its unprotected extremes—non-state actors pick up responsibility for doing so. In some ways this is desirable, because the state has certain powers we do not wish to see deployed against speech. But the state also has certain procedural safeguards which the community lacks.

The thing that everyone hates about cancel culture, after all, isn’t the idea that some statements merit social opprobrium. It’s the lack of process afforded to those accused of transgressing ill-defined lines, issues that the rule of law and due process are meant to address. Similarly, the constitutional prohibition on ex post facto laws does not apply to culture—what offends us about someone being fired for decades-old tweets is that it is illiberal to punish someone for doing something that was not, at the time of the act, a crime.

Subscribe now

Whatever its demerits, the old way of thinking about speech—Arkes’s way—did not suffer from a lack of ordering principles. Indeed, it rested on the notion that some speech can be regulated by reference to its effects on civil order, and its tendency to disrupt that civil order—to undermine the practice of the virtue of civility. In Chaplinsky, the Court excluded certain kinds of speech on this basis, noting that its disfavored categories form “no essential part of any exposition of ideas, and are of such slight social value as a step to truth that any benefit that may be derived from them is clearly outweighed by the social interest in order and morality” which they undermine.

Recall that group libel, like all criminal libels, is primarily justified by the state’s interest in keeping the peace. Illinois’s law, characteristically, prohibits speech “which is productive of breach of the peace or riots.” And the Court, in the Beauharnais opinion, notes that “the [Illinois] Supreme Court's characterization of the words prohibited by the statute as those ‘liable to cause violence and disorder’ paraphrases the traditional justification for punishing libels criminally, namely their ‘tendency to cause breach of the peace.’”

This idea—that speech can be proscribed because it threatens social peace—occupies an uncomfortable place in modern constitutional jurisprudence. The “tendency” phrase that the Court used in 1952 alludes to the “bad tendency” test for measuring when speech is dangerous enough to prohibit. That test was overridden by 1969’s Brandenburg v. Ohio, which undid literally hundreds of years of law to conclude that speech could only be constrained if it risked “imminent lawless action.” But cases like Beauharnais and Chaplinsky remain good law, and in principle preserve the idea that certain speech can be prohibited not just because it harms individuals (or groups), but because it disrupts public order as such.

The principle that these decisions embody offers us a firmer grounds for objecting to hate speech than the hurt feelings upon which Delgado tried to build his case. The problem with hate speech is not that it makes people feel bad; it is that it contravenes civility, and can thereby disrupt the civil order. This is true because handing out pamphlets that label certain groups rapists and murderers is liable to start a riot. But it is also true because civility is a necessary predicate of public life, which in turn is the foundation of a democracy.

This idea is a more bitter pill to swallow, though, than it may first seem. If you are willing to appeal to civility, after all, then you cannot stop at hate speech. You probably ought to extend it to disorderly behavior, obscenity, and the disruptive protests that amount to civil terrorism. Control of hate can be justified by reference to a socially conservative set of values that most advocates of hate speech laws are uncomfortable with or even openly hostile to.

I am not persuaded of the necessity of hate speech laws, even within the limited constitutional confines that apparently permit them. If we are to have group libel laws, I favor their appearance in name only, if for no other reason than that they are extremely hard to enforce. But I do think that there’s a strong argument that some egregious public expressions of hatred and intolerance imperils public order as such, and that some forms of regulation can be justified on those grounds—grounds firmer than appeals to hurt feelings. As I wrote several years ago in a report on hate crime:

Norms of respect regardless of difference are a precondition of the functioning of a polity like New York City. Hate crime enforcement establishes certain behavior as beyond the bounds of acceptable and therefore safeguards civic life. In one sense, this view is instrumental: prosecute hate crimes because doing so preserves civic peace; in another sense, it is intrinsic: prosecute hate crimes because their commission offends public decency more so than other crimes.

Hate crime enhancements are not hate speech laws; they regulate conduct, not expression. They do so by imposing an extra penalty on those who commit otherwise normal crimes motivated by bias. And they are justified in doing so because civic life is grounded in civility, and they contravene that virtue.

The idea that civility is an important part of democratic life does descriptively ground our historical objections to hate speech (in the form of “group libel” laws). And it should ground our decision-making about what speech is socially, and perhaps legally, impermissible. That we are no longer able to speak in the terms of civility perhaps explains the incoherence of contemporary speech discourse. But there is a history and tradition of doing so, and it would be good to recover it.

The event is this coming Tuesday at the Howard Theater. Doors open at 6. Tickets are only $25, so you can buy them and still have enough to pick up a couple of doses of fentanyl before the show. (Kidding! Kidding!)

You can get tickets here.

I don’t know how many of my readers are in D.C., but if you are, please show up and support us — gotta outweigh the Reason crowd. Plus, I think there will be some kind of after party? So if you’d like to meet me, now’s the chance.

This essay series has made three arguments about the War on Drugs. One is that it is a misnomer to identify one grand “War on Drugs,” stretching from Nixon to today. This is because, two, the War on Drugs came out of federal policy responses to several concrete drug problems—heroin, adolescent use, crack—rather than out of a consistent ideological response to drugs. Most importantly, three, the War on Drugs was a response to popular, grassroots outrage at those drug problems—an expression of the popular will which democracies are designed to channel.

What has received less focus up to this point is the actual effectiveness of the War on Drugs. Broadly speaking, the Reagan-Bush-Clinton drug war consisted of a dramatic increase in drug-related enforcement, supplemented by a substantial increase in prevention messaging, and by a government-led cultural backlash against drugs. Did these interventions actually work?

Subscribe now

Some of the most aggressive criticisms of the Drug War’s implementation are unmerited. Most significantly, critics routinely charge the War on Drugs with being the major cause of America’s uniquely high incarceration rate. The U.S. prison population rose from less than 200,000 in 1970 to over 1.5 million at its peak in the late 2000s.1 Locking up “low-level, non-violent” drug offenders, it is often claimed, was the major driver of this nearly seven-fold increase. For example, in her oft-cited book The New Jim Crow, Michelle Alexander declares that “in less than thirty years, the U.S. penal population exploded … with drug convictions accounting for a majority of the increase.”2

This is, simply, false. Drug offenders never accounted for a particularly large fraction of the U.S. prison population, peaking at about 22 percent of state prisoners in 1990.3 (Today the rate is 13 percent.)4 The drug offender share did spike in the early 1980s. But as John Pfaff—a Fordham law professor and ardent critic of mass incarceration—has shown, other categories of offenders grew too, such that drug-related incarceration explains only about 20 percent of the growth in the prison population between 1980 and the population’s peak in 2009. Most of that contribution, furthermore, comes in the 1980s. From 1990 onwards, drug offenses account for only 14% of prison growth, compared to 60% attributable to violent offenders.5

The oft-used construction “low-level, non-violent drug offenders,” furthermore, implies a relatively benign population sent away by the War on Drugs. The implication is that our nation’s jails are filled with people whose worst crime is a couple of puffs on a joint. Again, the reality is quite different. One analysis of prison populations circa 1997—near the peak of the Drug War—found that just 2% of federal and 6% of state offenders were “unambiguously low-level,” meaning that they had no prior convictions or arrests, were not involved in a “sophisticated drug group,” and did not use a gun in their crime. The remainder either had some criminal history or concurrent offense, were involved in a drug dealing enterprise (often at a “middle management” level), or used a gun in their offense.6 Most of the drug offenders locked up in 1997, in other words, weren’t innocents swept up by the War on Drugs—they were people with serious involvement in drug dealing or criminal life. Indeed, that shouldn’t surprise readers. The crack epidemic caused unusually high levels of violence, and so enforcement against it tended to sweep up very violent offenders.

But, given that the War on Drugs’ involvement in mass incarceration is overblown, what did it actually accomplish?

It seems to have dramatically reduced the social harms of the crack epidemic. Crack houses declined in prevalence, as did associated prostitution. Most importantly, the push to lock up members of violent drug gangs was likely a significant contributor to the great decline in crime in the 1990s and 2000s. The economist Steven Levitt once estimated that the decline of crack in general probably explained about 15 percent of that drop.7

On the question of drugs themselves, it seems like Americans, especially teenage Americans, really did change their minds about how dangerous drug use was. Gone were the days of cocaine paraphernalia on magazine covers. For example, high-school seniors (the group for which we have the most data) in 1979 were relatively sanguine about cocaine: only 32 percent said there was “great risk” in trying it. By 1994, that figure peaked at 57 percent. Support for drug-law reform also sputtered out. In 1977, 28 percent of Americans said marijuana should be legal, a 16-point gain over the preceding eight years. In 1985, though, support was back down to 23 percent, and it rose only barely to 25 percent in 1995.8 The dream of marijuana legalization was dead for a generation.

Initially, the War on Drugs also had a remarkable effect on the total number of people using drugs. The share of high-school seniors using any illicit drug peaked in 1979, at 54 percent. It then fell more or less continuously for the next decade, bottoming out in 1992 at 27 percent. The class of 1992, in other words, was half as likely as the class of 1979 to use illicit drugs.9 Similarly heartening trends obtained in the adult population. In 1979, there were an estimated 25 million illicit drug users, including about 4.7 million cocaine users; 4.1 million had ever used heroin. By 1992, those numbers had fallen to 12 million, 1.4 million, and 1.7 million respectively.10

But then, somewhat alarmingly, progress slowed, and in fact slightly reversed. By the year 1997, there were an estimated 14 million illicit drug users, including 1.8 million cocaine users, and 2 million lifetime heroin users.11 Even more alarmingly, the trends among high schoolers reversed sharply, as 42 percent of the class of 1997 reported using illicit drugs. Some of that inversion is explained by marijuana smoking, but 21 percent of 1997 seniors used illicit drugs other than marijuana—up from 15 percent five years before.12

Stalled progress was particularly obvious among “hardcore” users. In a 2000 report, the Office of National Drug Control Policy estimated that there had been about 3.9 million hardcore cocaine users and 900,000 hardcore heroin users in 1988. Those figures fell to 3.3 million and 700,000 by 1992. But then progress stopped or reversed, and by 2000 there were still 3.3 million hardcore cocaine users, and again 900,000 hardcore heroin users.13 And because heavy users consume most of the volume of drugs consumed, the estimated volume of drugs actually used was essentially unchanged between 1985 and 1995.14

As early as 1993, the drug czar’s office was sounding the alarm on slowing progress. In a section of its National Drug Control Strategy, it noted that “two distinct fronts are emerging in the war on drugs”—casual drug users and “hard-core addicted users,” who were more resistant to efforts to reduce their use. “Until we can succeed in making … drugs less readily available on our streets,” the report read, “at higher prices and at lower purity, it will be difficult to make continued headway against hard-core drug use.”15

Prices did not rise, and purity did not decline—indeed, the opposite happened. Between 1982 and 1992, the purity-adjusted price of cocaine fell by 75%, then leveled out. The purity-adjusted price of heroin fell by about two-thirds between 1990 and 2003, while purity actually rose.16 After two decades of the War on Drugs, in other words, drugs were much cheaper and more potent than they were when the War on Drugs began.

Subscribe now

One way to interpret this evidence is that the War on Drugs was remarkably successful at increasing the social opprobrium around drugs. Public education campaigns like the “Just Say No” movement plausibly played a role in convincing the general public that drugs were bad and harmful. As a consequence, more and more casual users chose sobriety.

But by the available evidence, the War on Drugs was far less successful at reducing use among the “hard core” of users, mostly meaning people addicted to their drug of choice. And because heavy users account for the majority of use—because one person who uses three times a day uses more than 20 people who use once a week—reducing casual use could only have a relatively small impact on the total level of drug use, and, by some measures, drug harm.

Reducing use was a major priority of the drug war program. As early as 1989, the drug czar’s office identified the “essence” of the problem as “use itself.” “The simple problem with drugs,” the 1989 National Drug Control Strategy claimed, “is painfully obvious: too many Americans still use them. And so the highest priority of our drug policy must be a stubborn determination further to reduce the overall level of drug use nationwide - experimental first use, ‘casual’ use, regular use, and addiction alike.”17

But of course, not all drug use is equally harmful. All drugs carry a risk of addiction—by definition—but people can and do use drugs casually without lasting harm. That’s obviously true of alcohol, but it’s also true of “hard” drugs like heroin.18 That doesn’t mean readers should rush out and score some junk. It’s very hard to know in advance if using drugs will lead to harm—that’s what makes them risky. At the same time, though, it means that not all drug users do equal amounts of harm to themselves and to others. And an exclusive focus on reducing overall use levels can obscure this fact.

On the one hand, there were probably real benefits to reducing the number of people—especially the number of teenagers—doing drugs casually. On the other, the drug war’s fixation on drug use stigmatized those who continued to use as moral failures to be “held accountable” for their choices.19 The “brain disease” model of addiction, popularized in a 1997 paper by then-director of the National Institute on Drug Abuse Alan Leshner, was in part designed to argue that drug use was a health issue, not a moral failing. As Leshner put it, “the gulf in implications between the ‘bad person’ view and the ‘chronic illness sufferer’ view is tremendous. As just one example, there are many people who believe that addicted individuals do not even deserve treatment. This stigma, and the underlying moralistic tone, is a significant overlay on all decisions that relate to drug use and drug users.”20

Attaching strong moral opprobrium to drug use, and aggressively enforcing against it, might have been justified if it changed hardcore users’ behavior. Tough love can be worth it if it saves lives. But the evidence from users, levels of use, and price all suggest it wasn’t. Why not? Why didn’t a million arrests a year appreciably change the use behavior of people in addiction? The answer, somewhat interestingly, comes down to economics.

How drug enforcement affects drug use is a deceptively simple question. Intuitively, if a drug dealer or a load of drugs is taken off the street, the quantity available goes down. If the quantity of drugs available goes down, Econ. 101 tells us, the price will go up.21 Less product available at higher prices should mean that, all else being equal, people will consume fewer drugs.

But not all drug enforcement affects prices equally. In a famous 1986 paper, drug policy scholars Mark A. R. Kleiman and Peter Reuter observed that the price of drugs rose substantially between points on the supply chain. The ten kilograms of opium needed to make one kilogram of heroin, for example, cost about $1,000 at the farm gate. When it enters the United States, that same kilogram costs about $200,000 per kilogram—most of the added price comes in the intermediary steps. As a result, even very large fluctuations in the farm-gate price—dropping to $100, or tripling to $3,000—would have only marginal effects on the final price of heroin on the street. Thus, Reuter and Kleiman argue, enforcement efforts targeting source countries or interdiction probably do very little to affect consumption.22

But by this logic, street-level enforcement should still have some effect on the consumption of drugs. How much, though, is a different question. Economists have a concept, “elasticity,” which refers in part to how much people will change their consumption of a good in response to a change in the price of that good. If the price of, for example, kale, goes up one percent, people will buy less kale. If the amount of kale consumed falls by more than one percent in response, the demand for kale is said to be “elastic.” If it falls by less than one, then demand is said to be “inelastic.” The less elastic demand for a good is, the bigger the change in the price has to be to reduce the quantity consumed. In drug-market terms, the less elastic demand for drugs is, the more enforcement is necessary to get a desired reduction in drug consumption.

A lot of research has tried to capture the elasticity of demand for drugs. A 2020 summary found that overall, demand was only minorly inelastic—for each 10 percent increase in the price of drugs, demand falls by about 9 percent. But that statistic obscures underlying differences. Specifically, studies of heavy users find that a 10 percent increase in the price of drugs will only reduce their use by around 3 or 4 percent.23 Some estimates go lower—one study that used data on arrestees, with urine tests as independent measures of their drug use, found that among heavy users a 10 percent increase in price reduces consumption of cocaine by 1.5 percent, and heroin by 1 percent.24

In other words, the way that people respond to drug enforcement depends on their habits. Casual users, less attached to their drug of choice, are prone to cutting back or not using altogether—just as they appear to have done during the 1990s. But heavy users—people who are addicted to drugs—face a significant cost from cutting back or quitting, whether it be the loss of euphoria or the suffering and risks involved in withdrawal.

That doesn’t mean enforcement can’t affect addicted users’ behavior—people suffering from drug addiction are not totally irrational, and they can and do respond to changing incentives.25 But it does mean that it takes much more enforcement to get an addict to change his or her level of behavior, relative to a non-addict.

Stated even more simply: it doesn’t take a lot of arrests and incarceration to get casual users to quit. But it does take a lot of enforcement to get the “hard core” to change their behavior. And while there in fact were very high levels of drug enforcement in the 1980s and 1990s, these levels do not seem to have been high enough to make an appreciable dent in the population whose drug problems were most acute and most harmful.

Subscribe now

Faced with these facts, one response is to say War on Drugs simply didn’t go far enough. If the level of enforcement in the 1990s wasn’t high enough to really reduce hardcore drug use, we simply need to do more. And, empirically, it is the case that super-tough crackdowns on drugs can curtail drug problems, as the Chinese Communist Party and the Taliban have both shown.26 But such crackdowns have costs too—in criminal-justice expenditures, and in lives and liberty lost. More to the point, it is unlikely that a liberal, democratic society like America’s would tolerate such a crackdown, particularly given the political unpopularity of the comparatively mild enforcement regime of the 1980s and 1990s. If the War on Drugs is so unpopular, imagine how unpopular a war modeled on the Taliban would be.

Another response would be to argue that drug enforcement isn’t worth the costs. In a very weak sense, this is true: enforcement could at least plausibly be reduced at the margins, as indeed it has been since its peaks in the 1990s. But the stronger conclusion—that cops should have no role in handling the drug problem—is something else entirely. As I argue elsewhere, there are clever enforcement strategies that overcome the problems identified in the last section. Policing drugs smarter, not harder, can have real impact on drug-related harms.

Although the War on Drugs stalled—predictably—in its efforts to create a “drug-free America,” it does not deserve its reputation as a destructive, racist failure. It deserves to be better remembered, in part because of what it was: a popular backlash against serious, community-wrecking problems with drugs that arose in the ‘70s and ‘80s. That same backlash probably substantially reduced the number of people using drugs. It helped turn the tide on the drug culture, and helped tamp down the worst excesses produced by that culture, from crappy magazines to crack houses. That it failed at its most important task is still a far weaker criticism than those so often leveled against it.

Demonizing the drug war has long been a key weapon in the arsenal of liberalizers. When we recognize it for what it was—a noble and reasonable, if fundamentally flawed, experiment—then that weapon should lose some of its power. The future of American drug culture isn’t, and shouldn’t be, a new War on Drugs. But it shouldn’t be a rejection of its ethos either, not properly understood.

Subscribe now

While Nancy Reagan was a committed drug warrior from the start, her husband took longer to put together a coherent strategy. That is not to say Ronald Reagan didn’t think about drugs, or wasn’t tough on them. At his second presidential news conference, when asked about “growing concern about the drug abuse problem,” Reagan called it “one of the gravest problems facing us internally in the United States.” But Reagan also seemed skeptical of drug interdiction, saying that “it’s like carrying water in a sieve.”

Rather, Reagan said, “it is my belief, firm belief, that the answer to the drug problem comes through winning over the users to the point that we take the customers away from the drugs, not take the drugs, necessarily … You don't let up on that. But it's far more effective if you take the customers away than if you try to take the drugs away from those who want to be customers.”1

Such demand-focused language sounds more like Kerlikowske than it does the popular conception of Reagan. And indeed, the focus on demand over supply would not last. By the time Reagan left office, the federal government was spending $4.5 billion annually on drug supply control efforts. That represented a three-fold increase, inflation-adjusted, over his first year. It was also more than double the $2.1 billion the federal government spent on demand reduction, a budget category essentially stagnant in inflation-adjusted terms until 1987.2 More money was paired with more action. Drug arrests reported to the FBI rose from about 500,000 in 1981 to over 800,000 in 1988.3 The fraction of new arrivals to state prisons convicted of drug charges rose from 8 percent to 30 percent in the same period.4

The change in tone, though, reflects the fact that Reagan’s early drug policy was often a response to the particular problems he faced, rather than an expression of a coherent ideology. For most of his first term, Reagan faced a series of supply-side drug problems. The rise of the brutal Medellin cartel in Colombia, and the ensuing wave of cocaine smuggling—first into Florida and then across the nation—prompted Reagan to go on the offensive, establishing task forces and seeking international cooperation.5 Domestically, when Reagan spoke about drugs—in his 1983 and 1984 State of the Union addresses, for example—he focused on their connection to crime, rather than drugs themselves. The most significant piece of federal drug-related legislation in Reagan’s first term was the Comprehensive Crime Control Act of 1984—a 400-page overhaul of the federal criminal justice system which, among other things, restored federal mandatory minimums for drugs and expanded the use of asset forfeiture in combatting drug crime.6

By his second term, though, drugs had become an issue of concern for Reagan—not just because of their connection to crime, or because of international conflict, but on their own merit. What changed was that Reagan, like Nixon before him, faced an acute crisis precipitated by a specific drug: crack.

Subscribe now

Cocaine, remember, saw its stock rise in the late ‘70s and early ‘80s. Peter Bourne was not the only physician lauding its qualities. In 1976, Harvard’s Lester Grinspoon and James Bakalar could comfortably write that “the most significant sociological fact about cocaine today is that it is rapidly attaining unofficial respectability in the same way as marihuana in the 1960s. It is accepted as a relatively innocuous stimulant, casually used by those who can afford it to brighten the day or evening.”7 This sense was reinforced by media coverage, which glamorized the “cocaine scene” and claimed that “cocaine is not addictive and causes no withdrawal symptoms,” citing doctors who happily backed up the claim.8

Use followed these cues. In 1976, about 6 percent of high-school seniors reported having used cocaine in the previous year. That rate doubled to 12 percent with the class of 1979, and peaked at 13 percent in 1985.9 In 1979, about 4.4 million Americans over 12 reported using cocaine on a monthly basis; the figure rose to 5.3 million by 1985.10 All of this use bred addiction: a nationwide cocaine hotline, opened in 1986, immediately began receiving thousands of calls a day. The callers were often high-class—lawyers, doctors, and engineers, many of them (contrary to popular perception) white. Their privilege didn’t protect them from reported insomnia, fatigue, headaches, sinus infection, depression, and irritability. Almost half claimed to have stolen to feed their habits.11

Just as for these men, the high of the cocaine culture was followed by a crash. For many, the precipitating event was the 1986 death of Len Bias, a University of Maryland basketball superstar. Bias was slated for NBA fame, and his sudden death from a cocaine overdose shocked a nation convinced that coke was basically benign. As one then-young reporter later told drug historian Jill Jonnes, “cocaine was still the perfect drug, a recreational drug that was not necessarily highly addictive. … Could cocaine really kill a six-foot-eight, two-hundred-pound guy?”12

Bias’s death, though, was just part of a broader transformation of the cocaine market. When it first hit the American scene, cocaine was a high-status, expensive drug. Low levels of supply kept prices high, restricting its use to the weekend parties of the glamorous classes. But as demand rose, and as the cartel infrastructure in Latin America expanded, supply increased, and price began to drop. In 1979, about 50 tons of cocaine entered the United States. Ten years later, the volume had risen to over 200 tons. In roughly the same period, prices fell over 75 percent. Cheap product facilitated heavier users’ transition from snorting cocaine to smoking it, which produces a faster and more intense high. Cocaine powder was boiled with ammonia or baking soda, creating a rock of “freebase” cocaine, which, when smoked, produced a cracking sound—thus the name.13

Crack spread quickly. In 1984, among a group of misdemeanor drug offenders in Manhattan, 42 percent tested positive for cocaine. Two years later, the rate had doubled to 83 percent. By June of 1986, Newsweek would claim that crack was in seventeen major cities and twenty-five states.14

At the same time that casual users were retreating from cocaine, then, would-be addicts were swept up by a wave of cheap, high-potency product. “Longtime addicts,” Jonnes writes, “whose other drug abuse had allowed some semblance of normal daily life were quickly lost to crack.”15 More addicting than powder cocaine, crack soon trapped its users in the cycle of binge and crash typical of stimulant addiction. Ethnographic accounts of crack-house life document how users spent most of their time either looking for or consuming “scotty” (as in Captain Kirk’s request “beam me up, Scotty!”). In between, their lives are portraits of dysfunction: frequent, destructive sexual relations, ruined families, a constant struggle to keep or find work.16

But what really set crack apart was the social devastation it brought. Crack houses were not merely places where people could get high; they were also a blight on the neighborhoods in which they sprang up. Women who found themselves trapped in the crack scene often traded, or were forced to trade, sex for drugs, and prostitution around crack houses was common.17 Crack use during pregnancy often resulted in small, fragile, sickly newborns—so-called “crack babies,” whom many feared would grow up permanently disabled. (Those fears turned out to be overstated, although children exposed to cocaine in utero are at risk for a variety of worse outcomes.)18 All of these problems were particularly acute within poor, black communities. One analysis, which measured the spread of crack using a composite of different measures, found that the drug’s rise between 1984 and 1989 explained roughly a third of the increase in low birth-weight black babies, and “much or all of the increase in Black child mortality and fetal death.”19

At the same time, crack markets were distinguished by their uniquely high levels of violence. Historically, street-level drug dealing had been dominated by organized crime rackets.20 But crack was different: disorganized street gangs became the main distributors, and engaged in bloody wars over the turf on which they sold.21 One scholarly analysis found that homicide rates among young, black men doubled after crack first appeared in a given city, and that that effect persisted nearly two decades later. By the year 2000, the data suggest, eight percent of all murders were still attributable to the long shadow of the early crack markets.22

Subscribe now

It should therefore not surprise readers—although it may anyway—to learn that black Americans began to vocally endorse a renewed War on Drugs. While the anti-drug movement of the 1970s had been, with some conspicuous exceptions, a majority white phenomenon, anti-drug activism in the era of crack was increasingly led by black Americans. By 1986, drug historian Emily Dufton writes, “minority activists were forming parent groups of their own in urban areas across America, and centering their fight on adolescent use of significantly harder drugs like crack, heroin, and PCP.”23 Indeed, the first of the “Just Say No" clubs was probably organized by Joan Brann, a black activist in Oakland, California, at the instigation of twelve-year-old Nomathemby Martini.24

Typical of this new, black movement was the Philadelphia anti-drug activist Herman Wrice. Raised in the predominantly black Mantua neighborhood, Wrice had a history of community organizing stretching back to the 1960s. Crack first came to his attention when students on the baseball team he coached started showing up high to games. Wrice and a group of other fathers marched on a nearby crack house with protest signs. Wrice—six foot four and armed with a sledgehammer—advanced on the house, only to find that its denizens had fled in terror; he and the other dads promptly boarded it up.25

What started as one protest became many. Organizing Mantua Against Drugs—which would eventually become the Philadelphia Anti-Drug Coalition—Wrice and other community members held all-night vigils on crack corners, protesting to demand their streets back. With the support of the Clinton Department of Justice, Wrice soon took his street vigil method national; by 1996, he claimed to have trained over 350 communities.26

Wrice was far from the only black anti-drug activist. Groups like Washington, D.C.’s Fairlawn Coalition or Detroit’s REACH organized similar anti-drug protests.27 Many of these groups were peopled and led by black residents, angry at the drug dealing that had taken over their neighborhood. Contrary to contemporary assumptions, many were happy to work with the police in the name of bringing peace and order to their communities. In Los Angeles, groups like the Urban League decided that police misconduct was a “lesser evil” compared to drug gangs, and endorsed more policing to get crack under control.28

This popular action was matched by an aggressive federal response to the crack crisis. In a September 1986 address to the nation, President and First Lady Reagan sounded the alarm about drug abuse, and particularly about “a new epidemic: smokable cocaine, otherwise known as crack … an explosively destructive and often lethal substance which is crushing its users.” Reagan announced that he would soon debut a plan for a “drug-free America.”29

A month later, Reagan signed the Anti-Drug Abuse Act of 1986, the signal legislation of the War on Drugs. The so-called “Len Bias Law” harshened federal drug-trafficking sentences and substantially increased funding for drug law enforcement.30 It also established mandatory minimum sentences for cocaine offenses, with a much smaller weight of crack versus powder cocaine—a 100-to-1 ratio—required to trigger the sentence.31 A major follow-up law, the Anti-Drug Abuse Act of 1988, further enhanced the drug-control system, created the federal “drug czar” office, and declared the goal of a “drug-free America” by 1995.32

The ADAA was in part an expression of collective discontent within black communities about the drug situation. One of the key leaders in passing the 1986 ADAA was Rep. Charles Rangel, the black Democrat who had then already represented Harlem for 15 years. Drugs were Rangel’s “signature issue,” because he believed that every problem his constituents faced was linked to the drugs destroying their community.33 Thanks partly to Rangel’s leadership, most members of the Congressional Black Caucus backed the ADAA.34

Subscribe now

Reagan’s most aggressive action against drugs really began only in the waning years of his administration. But his torch would be picked up by George H.W. Bush, arguably the president most committed to the drug war. In a famous televised address delivered a few months after he took office, Bush held up a bag of crack to millions of watching Americans and declared that “the war on drugs will be hard-won, neighborhood by neighborhood, block by block, child by child.”35

Under Bush—and Bill Clinton—the Drug War really took on the enforcement-driven character we associate with it today. In Reagan’s final year in office, the federal government spent about $2 billion per year on criminal-justice-related drug expenditures (arrest, prosecution, incarceration, etc.), and another $1.2 billion on interdiction. By the time Bush left office, in 1993, those figures were $5.7 billion and $2.1 billion. In 1998, midway through Clinton’s second term, it was $8.2 billion and $2.4 billion—more than $10 billion per year for fighting the War on Drugs.36 Drug arrests rose, and by 1994 police departments were collectively reporting over a million per year.37 In 1985, there were fewer than 40,000 people in state correctional facilities for drug offenses; by 1995, there were 225,000.38

Today, the Drug War is unpopular. But in its heyday it wasn’t. In 1988, a sizable majority of Americans said government drug policy should emphasize stopping drug importation and arresting drug dealers.39 A poll fielded days after Bush’s crack speech found that 64 percent of Americans called drugs “the nation’s most important problem.” A similar share reported that they favored Bush’s strategy for fighting drugs, and believed it would reduce the drug abuse problem.40 In many ways, this optimism remained: by the year 2000, 47 percent of Americans said they believed the nation had made progress in coping with the problem of illegal drugs, while just 29 percent believe they had lost ground. These were, respectively, the highest and third-lowest shares on record.41

But were they right? Was the War on Drugs actually getting the drug problem under control?

In 1979, the federal government conducted a survey of Americans’ drug use habits. They asked, among other things, about when respondents had first tried drugs. The results were quite startling. Rates of drug initiation, the responses imply, were quite low prior to 1960. Only about 2.4 million Americans alive in 1979 first tried any of the major illegal drugs (marijuana, cocaine, heroin, or hallucinogens) in 1960 or before. But 1.8 million tried drugs in the years 1961 to 1964—almost as many as the preceding several decades. The number grows from there. Between 1965 and 1968, 7.8 million first tried drugs; 1969 to 1972, 16.5 million; and 1973 to 1976, 17.8 million. As many as 5 million people probably tried drugs for the first time in each of 1977 and 1978.1

An analysis of retrospective data like this is necessarily imperfect.2 Nonetheless, it corroborates the widely held belief that in the 1960s and 1970s, many more Americans started doing many more drugs. That increase in drug use, furthermore, reflected increasing social comfort with the use of drugs. Thus one of the problems with assuming a continuous drug war from Nixon to Reagan: through much of the 1970s, policymakers and culture declared a ceasefire.

For decades, drugs were shunned. But in the 1970s, they were suddenly everywhere. Whether it was the Beatles’ psychedelia or marijuana songs like Brewer & Shipley’s “One Toke Over the Line,” drugs were all over the radio. Pot-themed publications sprang up. High Times, the most famous, at one point sold over 400,000 copies a month, and had a 100-person staff.3 By 1978, the New York Times was breathlessly reporting on head shops that sold paraphernalia made for kids, like “a baby bottle fitted with both a nipple and a hashish pipe and a felt-tipped pen that allows a surreptitious snort of cocaine.”4

The cultural interest went beyond pot. While the heroin epidemic was still visible enough to make it unattractive, cocaine was rapidly gaining popularity. A 1971 article in Newsweek—circulation 2.6 million and rising5—called cocaine “the status symbol of the American middle-class pothead,” mentioned the fashion of wearing a spoon around the neck, and included a “co-ed” claiming that “orgasms go better with coke.” The following year, glowing coverage in Rolling Stone blamed cocaine’s prohibition on “absurd laws about harmless drugs.”6

Cultural change begat political activism. Pro-liberalization groups popped up throughout the country—marijuana historian Emily Dufton7 identifies not just major groups like the National Organization for the Reform of Marijuana Laws (NORML) but local outfits like Atlanta’s Committee Against Marijuana Prohibition (CAMP) or Beloit College’s Student Association for the Study of Hallucinogens (STASH).8 These activists moved public opinion on marijuana: support for legalization rose from 12 percent in 1969 to a peak of 28 percent in 1977, a remarkable shift in public opinion in such a short time.9

Even more remarkably, policymakers actually listened to them. Between 1973 and 1978, a dozen states, collectively home to more than a third of the population, passed laws removing criminal penalties for possession of up to an ounce of marijuana.10 When Jimmy Carter became President, he made his chief drug policy advisor Dr. Peter Bourne, a physician who had once called cocaine “probably the most benign of illicit drugs currently in widespread use” and argued that the case for cocaine legalization was at least as strong as that for marijuana.11 Under the influence of both Bourne and NORML’s Keith Stroup, Carter endorsed marijuana decriminalization nationwide in 1977.12

That, of course, did not come to pass. And Bourne eventually resigned after it was reported that he had written a prescription for quaaludes for an aide under a false name and snorted a line of coke at a NORML party.13 But the fact that the Carter White House embraced marijuana decriminalization shows that the “War on Drugs” was not a continuous campaign, but fought in stops and starts for its first decade. More to the point, it shows the ways in which drugs and drug use became more normal in the ‘70s: from the magazines to the campuses to the White House, millions of Americans were saying yes to drugs.

Which is exactly what had Keith Schuchard so worried.

Subscribe now

Marsha “Keith” Mannat Schuchard was an unlikely leader of an anti-drug movement. She was a liberal Democrat, not a fire-breathing conservative. She had tried pot while she and her husband were studying for their PhDs in literature. But that didn’t stop Schuchard’s surprise one warm August evening, just after her eldest daughter’s birthday party. The party, Schuchard reported, had been odd, the attendees—mostly seventh and eighth graders—“bleary-eyed” and “incoherent.” When she and her husband surveyed the results of the party, they found “marijuana butts, plastic bags with dope remnants, homemade roach clips, cans of malt liquor, and pop wine bottles” strewn across the lawn.14

Schuchard reached out to the parents of the party’s attendees. From those conversations came a meeting, at which attendees expressed how out of control they felt their kids’ drug use had become. They formed a group—their kids derisively labeled it the “Nosy Parents Association”—to work together to discourage drug use.15

What started as a suburban parents’ association soon became a movement. With the aid of Robert DuPont, the Nixon drug advisor, and Georgia State health professor Buddy Gleaton, Schuchard created the Parents’ Resource Institute for Drug Education, or PRIDE. By 1979, PRIDE had a nationally circulating newsletter. By 1980, Dufton writes, “there were hundreds of Schuchard-inspired groups across the country, calling themselves Parents Who Care, Parents Alert, United Parents of America and a host of other names.”16 Sue Rusche, another early Parent Movement leader, said that within the first few years of the movement there were some 3,000 groups nationwide.

“There was just this explosion of anger, on the part of parents that were living then and raising kids,” Rusche said. Parents were asking “how can this be happening in my community? And I'm gonna make it stop.”17

Parents’ concerns went beyond illegal drugs. On May 3, 1980, 13-year-old Cari Lightner was struck and killed by a drunk driver. Police later told Cari’s mother, Candy, that the man who had killed her daughter had three prior drunk driving arrests. Outraged, Candy joined together with some friends to launch MADD: Mothers Against Drunk Driving. In the next decade, thanks largely to MADD’s advocacy, states would pass hundreds of drunk-driving laws.18

It wasn’t just white, middle-class parents like Schuchard who were worried about drugs in their communities. Black parents, too, were speaking out. In New York, Harlem’s black community leaders had already rallied a “black silent majority” around support for Governor Nelson Rockefeller’s punitive drug-law reforms.19 Detroit’s first black Mayor, Coleman Young, “brought the majority-black audience of more than two thousand people to its feet” by declaring that “I issue an open warning right now to all dope pushers, to all rip-off artists, to all muggers: it’s time to leave Detroit.”20 And in Washington, D.C., it was black leaders who led the fight against marijuana decriminalization, as a majority of black residents opposed it against a white majority that supported it.21

The Parents Movement, as it would come to be called, transformed the conversation around drug use. They did so, Dufton argues, by changing the locus of concern from the rights of adults—the right to consume drugs, especially marijuana, without government interference—to the needs of children.22 The emergence of MADD, with its emphasis on mothers’ outrage, reflects the same tendency: a sense that substance use imperils children, and the rights of adults should be curtailed to reduce that risk.

It might seem easy to see parents like Schuchard as unhip moms and dads who were freaking out over nothing. But the reality is that, by the time they started noticing, teen drug use was a large and growing problem in America. In 1976—the first year the government asked—48 percent of graduating seniors reported using any illicit drug in the past year. That figure would peak at 54 percent in 1979. Many of these were “just” smoking marijuana. But fully a quarter of seniors reported using illicit drugs other than marijuana in 1976, and more than a third in 1981.23

To be sure, lots of kids experiment with drugs. For many, those experiences are fine, even pleasant. But many also get hurt. A teen who smokes and drives can die in a car crash, or oversleep and misses a crucial test. Or they can get addicted, leading at least to debilitating symptoms, and in many cases to ongoing mental health problems—a particular risk in exposing the developing teenage brain to drugs. One does not have to be a breathless square to recognize that it's not good for kids to do drugs, and it’s not good for a culture to glamorize drug use if it means kids getting high.

More to the point, understanding the War on Drugs means understanding that, by the late 1970s, American culture really was glamorizing drug use, often to harmful effect. The history of the parent movement—rarely mentioned in contemporary polemics against the Drug War—captures how the 1980s backlash against drugs was the result of a grassroots movement, grounded in a legitimate concern about real and substantial problems.

In fact, without the parent movement, it’s hard to say if there would have been a renewed War on Drugs at all. Because by 1980, a champion of the movement was about to enter the White House.

Parent movement groups “were warmly received by the Reagan administration,” drug historian David Musto writes.24 In particular, the movement found a close ally in First Lady Nancy Reagan. Even before she was in the White House, Mrs. Reagan worked closely with movement leaders.25 The relationship persisted when her husband entered office, as Mrs. Reagan drew staff from the National Family Partnership, a parent movement umbrella organization. 40 NFP members were present, too, at a November 1981 meeting at which the first lady announced she would focus her time in the White House on preventing adolescent drug abuse.26

That campaign eventually begat Mrs. Reagan’s now-famous slogan, “Just Say No.” The phrase, she claimed, came from a meeting with a group of children in Oakland, California. “I was asked,” the first lady recalled in 1986, “what to do if they were offered drugs. And I answered, ‘just say no.’”27

“Just Say No” today is often recalled with derision, an almost hopelessly naïve slogan for a more foolish time. But that unflattering image belies Mrs. Reagan’s legacy as a prevention activist, an extension of the parent movement. By the end of Reagan’s second term, the first lady’s work had led to the founding of over 12,000 “Just Say No” clubs and gotten more than five million people to attend Just Say No marches in 700 cities.28 “Just Say No” may have been a joke to some, but to others it was a serious slogan for those fed up with drugs.

Subscribe now

The declaration of the War on Drugs is usually dated to an otherwise unremarkable press conference in June of 1971. President Richard Nixon, emerging from a two-hour meeting with advisors, did not actually use the phrase “war on drugs.” But he did tell the press that he was prepared to “wage a new, all-out offensive” against “America's public enemy number one”: drug abuse. Nixon promised a “worldwide” effort, and announced the formation of a new White House office responsible for coordinating drug policy across the federal government.1

Nixon, of course, did not invent drug-control policy. The first major federal drug-control law, the Harrison Narcotics Tax Act, was passed in 1914. Starting in 1930, the Federal Bureau of Narcotics waged an enforcement-focused campaign for almost four decades under the leadership of its infamous director, Harry Anslinger. Months before the press conference, Nixon had signed the Comprehensive Drug Abuse Prevention and Control Act of 1970, which created the substance scheduling system that remains the primary drug law of the land.

Nixon’s announcement, then, did not really inaugurate a transformation of drug policy per se. Rather, it changed its focus to the drug problem of the moment: heroin.

Subscribe now

Heroin has been in America a long time. In the 19th and early 20th century, it was available over the counter. After prohibition, it remained popular in some milieus, especially the jazz clubs of the 1930s and 1940s.2 But the 1960s ushered in a full-blown heroin epidemic. In New York, the number of narcotic addicts doubled between the early 1960s and the early 1970s.3 In Atlanta and Boston, the number rose tenfold.4 In Philadelphia, the number of heroin overdose deaths rose from 5 per year prior to 1962 to 170 by 1970. “Jumping death rates,” one physician wrote in 1970, “are found from New England to Miami, Fla, from New Orleans to Seattle.” He further noted an alarming increase in youth use; heroin was then the leading cause of death in New Yorkers ages 15 to 35.5

Many were worried that the withdrawal from Vietnam would compound these problems. Heroin use had become widespread among GIs—a 1970 survey found that about 12 percent had tried it, and half of those were still using.6 As thousands of men returned home each month, one in twenty was found to have used heroin just before departure.7 An already simmering crisis might soon hit a boil as millions of vets returned still hooked on dope.8

Nixon’s War on Drugs, in other words, was not motivated purely by an irrational dislike of drugs. Nixon of course did hate drugs, especially marijuana, with a passion out of step with the consensus among elites, then and now. But his policy agenda was responsive to a real and substantial drug epidemic, one which merited a proportional government response.

This response, furthermore, belied the now-popular view of Nixon as an all-enforcement drug warrior. Certainly, Nixon saw interdiction and policing as important components of policy, as communicated by both his rhetoric and, in 1973, by the establishment of the Drug Enforcement Agency. Federal narcotics arrests doubled, from about 8,000 to about 16,000, between 1969 and 1972.9 And Nixon also made drugs part of his international diplomatic efforts, for example in convincing the Turks to shut down opium poppy production in their country, which by 1971 was the principal source of roughly two-thirds of U.S.-consumed heroin.10

But, by comparison to what came before and after, the Nixon approach was a model of progressive drug policy. While he expanded enforcement spending, he also oversaw the repeal of most federal mandatory minimums for drug crimes.11 Nixon was also very clear that enforcement was not the only tool in his arsenal: “It has been a common oversimplification,” he told Congress in 1971, “to consider narcotics addiction, or drug abuse, to be a law enforcement problem alone.”12 Prior to the 1970s, federal drug policy had indeed focused entirely on enforcement. But Robert DuPont, who served as Nixon’s second drug advisor, has written that under Nixon enforcement was “augmented by an entirely new and massive commitment to prevention, intervention and treatment.”13

The numbers corroborate this claim. In 1970, the federal government spent nearly $60 million on “demand reduction”—treatment and prevention programming—compared to $53 million on “supply reduction” (interdiction and enforcement). By 1973, Nixon’s last full year in office, his administration was spending $466 million on demand, versus just $214 million on supply.14 With this money, the administration bragged of educating 440,000 students, teachers, and community leaders, and offering treatment to over 20,000 returning soldiers.15

This focus on demand was thanks not so much to a coherent theory of drug policy as to a remarkable new treatment for opioid addiction: methadone. Like heroin, methadone is an opioid. Unlike opium, it has a much lower potential to induce a high, especially when taken orally. In a landmark 1965 study, doctors Vincent Dole and Marie Nyswander showed that methadone administered to heroin addicts relieved withdrawal without inducing a high. “With this medication, and a comprehensive program of rehabilitation, patients have shown marked improvement,” the pair wrote. “They have returned to school, obtained jobs, and have become reconciled with their families.”16

Methadone maintenance soon drew politicians’ attention. In 1965, Dole presented the idea to New York City’s hospital commissioner. By 1969, there were almost 2,000 New Yorkers enrolled in methadone maintenance programs; by 1970 there were 20,000. As New York enrollees showed major gains, programs were set up across at least 23 cities.17

One of the young men working in Dole’s lab was Dr. Jerome Jaffe—the same man whom Nixon would, in the infamous 1971 address, name to coordinate federal drug control policy. Thanks in large part to Jaffe, historian Jill Jonnes writes, “the Nixon administration was responsible for opening the first legal opiate clinics since the twenties.” By the end of Nixon’s administration, the nation had some eighty-thousand methadone program slots—enough to warrant the closure of the national drug treatment hospital at Lexington.18

Subscribe now

Critics of Nixon’s War on Drugs, looking to link it to later efforts, often cite an interview with Nixon aide John Ehrlichman, who in 1994 allegedly told a journalist, “we knew we couldn’t make it illegal to be either against the war or black, but by getting the public to associate the hippies with marijuana and blacks with heroin, and then criminalizing both heavily, we could disrupt those communities. … Did we know we were lying about the drugs? Of course we did.”19 For opponents of the War on Drugs—and for the public that listens to those opponents—Ehrlichman’s confession is taken as final proof of what they have long suspected: Nixon’s drug policy was mostly a pretext to hurt black people and the left.

Usually left out is that Ehrlichman was such a liar that he was actually convicted of perjury in connection to Watergate, and that he held grudge against Nixon for—in Ehrlichman’s view—sucking him into the Watergate debacle that led to Ehrlichman’s imprisonment.20 A reliable source, he was not.

Beyond Ehrlichman’s credibility problems, even a passing review of the history suggests that we should identify Nixon’s drug war with the broader social reform agenda of the late 1960s. The phrase “War on Drugs,” after all, follows the same formulation as Lyndon Johnson’s “War on Poverty” and “War on Crime.” All three were all-of-government efforts to eradicate social ills: the essence of liberal meliorism, not its antithesis.

It's not just that the standard story of the War on Drugs is wrong, nor that Nixon doesn’t deserve much of the criticism he receives. It’s that Nixon’s War on Drugs was an expression of the progressive ethos of that era, rather than a rejection of it. Nixon—the most conservative president between Hoover and Reagan—still prosecuted a drug control campaign that was fairly liberal by today’s standards.

Subscribe now

As a result, I put a great deal of work into researching essays that never saw the light of day. While I have repurposed some of that work, I have long wanted to be able to reference other output from that fellowship in subsequent writings, only to remember that it has never been formally published anywhere.

Parts II, III, IV, and V are available at their respective links.

This series adapts an essay, never previously published, which offers a revisionist history of the War on Drugs. I am obliged to thank the Fund for American Studies for helping me write it.

This is the first installment. There will be five in total, released at a relatively fast tempo. Expect part II, on the Nixon administration, soon.

In May of 2009, just a few months after President Barack Obama took office, Gil Kerlikowske was installed as the sixth director of the Office of National Drug Control Policy. Often called the nation’s “Drug Czar,” ONDCP’s director oversees the federal government’s sprawling, often unwieldy efforts to control drugs and their problems. The Director also helps set the White House’s tone on drugs, which has historically been a firm one. William Bennett, the first director of ONDCP, once remarked that he had no moral problem with beheading street dealers.1 And when Bill Clinton wanted to prove his tough-on-drugs bona fides, he selected Barry McCaffrey, a four-star general decorated for his service in the Gulf War.

Which is why it was so surprising when, just a few days after he took office, Kerlikowske called the War on Drugs a failure.

“In the grand scheme, it has not been successful," Kerlikowske said in an interview with the Associated Press. "Forty years later, the concern about drugs and drug problems is, if anything, magnified, intensified."2 He told the Wall Street Journal that he wanted to dump the phrase “War on Drugs” altogether, explaining that “regardless of how you try to explain to people it's a 'war on drugs' or a 'war on a product,' people see a war as a war on them.”3 This rhetorical commitment persisted. In the 2013 National Drug Control Strategy, for example, Kerlikowske would write proudly of how the Obama administration had rejected a “law enforcement-only ‘war on drugs.’”4

Kerlikowske’s comments followed the Obama administration’s commitment to moving away from law and order and towards public health. But they also reflected the way in which the phrase “War on Drugs” has collapsed in popularity. In a 2012 poll by anti-drug-war magazine Reason, 80 percent of Americans said the War on Drugs had failed.5 That’s remained true since Obama left office. in a 2021 poll run by the American Civil Liberties Union, 83 percent of Americans said that they believed the War on Drugs had failed; 65 percent supported “ending” it.6

Why don’t Americans like the War on Drugs? To be fair, the standard picture of it is quite unappealing. The War on Drugs, the story goes, was started by Richard Nixon, primarily—John Ehrlichman once claimed—as a way to turn the American public against black people and hippies. This “moral panic” over drugs, which were never really that big of a problem, persisted, driving an insanely harsh backlash. Over the next fifty years, the United States spent a trillion dollars to lock up millions of people, most of them black, for the crime of possessing drugs. In so doing, they went to war on everyday Americans, destroyed communities, and created America’s mass-incarceration problem without making the supposed drug problem any better.

This version of events serves a political purpose. Popular discontent with the War on Drugs—the sense of its basic illegitimacy—is a major driver of support for drug liberalization. The War on Drugs has become a bogey-man, the thing to which reformers point as the only alternative to their preferred way forward. How, after all, can a right-thinking person not want to move away from the horror that was the Drug War—and therefore, presumably, make peace with drugs?

It matters, then, that we get the story of the War on Drugs right. And the standard story—the story you have probably been told over and over again—is quite simply wrong. It is, in fact, profoundly, deliberately misleading.

Subscribe now

Richard Nixon, for example, may have declared drugs “America’s public enemy number one.” But he did so in response to an acute heroin crisis, not to score cheap votes in the South. Nixon also spent more on drug treatment than enforcement every year after, and pioneered the use of methadone maintenance treatment.

Despite what critics claim, there is no fifty-year straight line from Nixon to Reagan’s drug war. Drug policy actually liberalized throughout the United States in the mid-to-late 1970s, reflecting and reinforcing an enormous surge in drug use, particularly among teenagers. It was this surge that first prompted a massive grassroots parent movement—ignored and forgotten by most modern drug-war historians—to demand government do something.

Even then, it took the crack epidemic—one of the most devastating drug crises of the 20^th century—to spur real, aggressive action. When that action came, it too was supported by a popular, grassroots movement, one often led by the very same black Americans the drug war ostensibly was designed to oppress. And, at least at first, this effort produced real results, as the number of drug users plummeted and neighborhoods, once blighted by crack, came back from the brink.

The War on Drugs, in short, deserves to be remembered for what it was: a popular reform movement targeting a major social problem. That popular movement, furthermore, did substantial work to fix the harms an uncontrolled drug culture had wrought.

At the same time, though, the War on Drugs does not deserve unqualified praise. The reality is that by the late 1990s, progress in the “war” had stalled. And that stalling was entirely predictable, a function of the inherent difficulty of shaping demand for addictive substances. After a certain point, it turns out, it quickly becomes very hard to police your way out of a drug crisis.

The War on Drugs of the 1980s and 1990s is not the right way forward for our current drug problems. But a correct understanding of the it should give us a sense that culture makes a difference: that it does actually matter, despite what you may have heard, whether or not we as a society are willing to “just say no.”

Subscribe now