Day: August 5, 2025

Ever since President Trump launched his protectionist tariff agenda at the outset of his administration, economists have offered a common plea to other countries: do not retaliate. This wisdom must often contest with the impulsive political pressures of trade wars, and yet it is the soundest economic path. Politicians often believe that retaliatory tariffs place a burden on the foreign country that picked the fight. Retaliatory tariffs tax your own population by burdening them with higher prices and fewer consumer choices. The actual result, however, is akin to imposing a blockade on your own ports to spite an adversary who chose to do the same to their ports. This strategy seldom ends the trade war though, because the bulk of the resulting economic pain is self-inflicted.

Trump’s early targets largely ignored the counsel of economists to forego retaliation. Canada, then in the midst of an election campaign, imposed countervailing duties on American goods as its Liberal government sought to project an image of strength in the face of international bullying. Trump responded by doubling down. The situation with China rapidly devolved into a retaliatory standoff with threatened US rates on Chinese imports reaching 145 percent, and China’s countermeasures increasing to 125 percent. Both sides de-escalated from this brink in May, but only partially. The US now imposes a 30 percent tariff on most Chinese imports, whereas China now taxes American goods at 10 percent—substantial hikes over the single-digit rates that prevailed in the 2010s.

Until last week, it appeared that the 27-member European Union was about to join the retaliatory trade war. With Trump threatening a 30 percent tariff on European goods, EU leaders floated the idea of invoking their “Anti-Coercion Instrument” (ACI)—an aggressive suite of retaliatory tariffs, export and import controls on the US, intellectual property restrictions, and limitations on American investors in Europe. Suddenly, on July 28, Trump appeared to pull a rabbit out of his hat by announcing a sweeping trade deal. According to the White House announcement, the EU would accept a 15 percent US tariff on European goods, or half the threatened rate, but also significantly higher than the average 1.47 percent duty that existed before Trump started his trade war. Pre-Trump EU rates on American goods were even lower, sitting at about 1.35 percent before Trump’s term, but the White House now claims that Europe will further reduce these mostly vestigial levies on a small number of products.

The US-EU announcement sent shockwaves around the political press. Trump declared “victory” over the Europeans. French officials raged at the outcome, calling on Brussels to reverse course and deploy the retaliatory ACI. Other European leaders seemed resigned to the fate of accepting higher American taxes on their exports, calling it the least-bad outcome. On the American side of the debate, some tariff-skeptics reacted with surprise, even suggesting that Trump’s tactics had upended conventional wisdom about trade wars. Oren Cass, the protectionist lawyer who runs the “New Right” think tank American Compass, appeared to be confused by the development even as he celebrated. Cass declared on a podcast that the EU’s unexpected turn away from retaliation somehow disproved all of the economists who had been pleading with the EU to resist retaliation.

Surprisingly few commentators have paused to examine the underlying strategy that led to the US-EU tariff de-escalation. Almost all sides of the debate have accordingly missed a subtler lesson, one related to legal and institutional dynamics at play.

Before digging deeper, it warrants mention that the EU “trade deal,” like the others that Trump has announced to date, is not a conventional trade agreement between nations. Neither Congress nor its European counterpart has voted on the specific terms, which may not even exist beyond the verbal pledges that Trump and European Commission President Ursula von der Leyen made in a meeting on the day of the announcement. There are no publicly available documents containing the text of this “deal” or any other that Trump claims to have secured. Trump and his counterparts have not even formally signed anything that would legally bind them to their terms. The EU deal is at best a framework of loose commitments that may or may not be formalized at some future date.

And therein lies the key to understanding why some nations have opted against a retaliatory strategy. The tariff agenda sits on an extremely shaky legal and political foundation. Tariffs remain unpopular in the United States, with recent polling showing that American voters oppose them at a margin of almost 2-to-1. The EU and other countries might accordingly believe that their best strategy is to wait Trump out until a future election outcome reverses his trade war.

More importantly, Trump has eschewed even conventional US tariff statutes when implementing his current measures. Up until this spring, most US tariffs followed the procedures laid out in obscure federal statutes: the Trade Expansion Act of 1962, the Trade Act of 1974, and a handful of other measures. These laws contain clauses that allow the president to impose temporary tariffs on specific goods or countries if certain stipulations are met, such as a national security need, or evidence that another country has been “dumping” goods into the United States or trying to injure its industries. 

The rationales behind these clauses are often politicized, but they nonetheless contain specific statutory requirements to invoke. Usually, that means the White House must formally investigate whether the conditions have been met and submit its findings to a regulatory review process. Congress also has formal oversight powers to review any tariffs enacted under these clauses, and may vote to reverse the president’s decision. Most presidents have made periodic use of these clauses to impose limited tariff measures, including Trump’s tariffs on China from his first term. But all of these measures have taken place under relatively strict statutory processes and review requirements.

Not so with Trump’s current round of tariffs, which are far more expansive than anything he attempted in his first term. Instead of using the aforementioned trade statutes, Trump has claimed the power to impose tariffs by unilateral executive order under the International Emergency Economic Powers Act (IEEPA) of 1977. By using IEEPA, Trump believes he can enact new tariffs for almost any reason he wants. He has accordingly set new tariff rates by posting letters to foreign leaders on Truth Social, or by announcing the terms of an “deal” that he made in an informal conversation with a foreign head of state.

The biggest risk to Trump right now is that the appellate and Supreme Courts both strike down his IEEPA tariff orders, and with them every “deal” he has secured to date.

There’s a huge constitutional problem with Trump’s use of IEEPA, though. This statute does not authorize the president to impose tariffs. It does not even contain the word “tariff” or any of its common synonyms. Trump has merely inferred that he has this power through a tendentious interpretation of IEEPA’s “emergency” provisions.

Trump’s use of IEEPA now faces a serious challenge in the federal court system. Earlier this year, the US Court of International Trade ruled that Trump misused IEEPA. The ruling is under a procedural pause as the case makes its way through the federal appellate system, but the clock is ticking. A federal appellate court will likely rule on the case in August, which in turn will probably fast-track it to the Supreme Court in its next term. If the courts continue to rule against Trump, his entire tariff agenda—including every single “deal” made under his IEEPA orders—could be voided in an instant.

Although Trump is fighting to have the trade court’s ruling overturned, the underlying legal issues present a tough sell for the White House’s lawyers. In addition to Trump’s strained interpretation of the IEEPA statute, he faces a constitutional obstacle. Article I, Section 8 of the US Constitution gives Congress the exclusive authority to decide tariffs and tariff rates under its tax and commerce powers. 

Previous court rulings have upheld laws wherein Congress authorized the president to modify tariff rates under certain conditions, such as the aforementioned Trade Acts of 1962 and 1974. But they’ve never authorized the president to simply set the rates himself for any reason he wants. The long-established “nondelegation doctrine” requires Congress to set an “intelligible principle” for the executive branch to exercise a delegated power, and the parallel “major questions doctrine” bars executive branch agencies from deciding policy questions of great economic and political significance on their own without express congressional authorization. The Court of International Trade cited both doctrines in its determination that Trump abused IEEPA to impose tariffs. A majority of justices on the current US Supreme Court have also ruled against the executive branch in cases involving similar questions, such as West Virginia v. EPA in 2022 and in their rulings against President Joe Biden’s attempt to forgive federal student loans without congressional authorization (a detailed explanation of these legal arguments may be found in the amicus briefs filed in relation to the case).

Returning to the international arena, Trump may be playing with fewer cards than he realizes when he attempts to negotiate tariff deals. The EU and other governments know that time is on their side with both the tariff lawsuits and the US political climate. And that recognition changes the dynamics of their strategies by shifting them away from retaliation. Consider the possible scenarios.

The biggest risk to Trump right now is that the appellate and Supreme Courts both strike down his IEEPA tariff orders, and with them every “deal” he has secured to date. Courtroom outcomes are never guaranteed, and they tend to operate on the judges’ own schedules, but this outcome could conceivably occur in a matter of months. If that’s the case, then the EU has every incentive to take the 15 percent tariff “deal” and just wait it out until the courts make their determination.

Suppose that the Supreme Court punts on the case for procedural reasons, or finds a rationale for upholding the tariff policies. Even in this scenario, waiting may still be the more prudent strategy for the EU and other trading partners abroad. They may be banking on Trump losing one or both houses of Congress in the 2026 midterms, as often happens to the incumbent president’s party. If the Democrats regain the majority, they will almost certainly remove a procedural rule that Republican Speaker Mike Johnson imposed back in April to prevent floor challenges to Trump’s IEEPA tariffs (Johnson adopted this rule because he currently lacks the votes to defeat a direct challenge if a handful of free-trade Republicans break ranks and vote with the Democrats). An opposition Congress would make it more difficult for Trump to enact tariffs by executive decree, thereby limiting his ability to use them as leverage for his “deals.”

A third scenario could play out in the 2028 election, where both political parties will be nominating new candidates due to Trump’s term limits. On the unlikely chance that the current IEEPA tariffs are still in place in 2028, their survival into the next administration would come down to a simple pen stroke. Since Trump enacted his tariffs by unilateral executive order, they could also be rescinded by any future president at will. Here, the absence of any codification of Trump’s tariff orders and “deals” into law becomes their main future vulnerability. If tariffs remain politically unpopular, Trump’s successor will face growing pressure to rescind them, no matter the party that wins.

From the EU’s perspective, these scenarios amount to a surprisingly short timeframe. Trump’s tariffs could be stricken down in court in a matter of months, or they could collide with mounting political opposition in 1.5 to 3.5 years.

Under these conditions, the advantage goes to the side with greater patience. Risky strategies such as the EU’s retaliatory ACI, or retaliatory tariffs in general, become less appealing if you know there’s a good chance that Trump will be stripped of his claimed tariff powers in the near future. It may mean enduring a little short-term pain from accepting the higher tariffs of a Trump “deal,” but in the long run, all of Trump’s negotiating cards have an expiration date. 

America is still reeling from the coronavirus pandemic that broke out in 2020. Not only was it one of the most deadly health incidents in our history, but the strategies imposed by central planners to contain its spread also inflicted countless costs on everything from the economy and education to social life itself. Stephen Macedo, an author of a recent book evaluating the pandemic’s aftermath, joins Law & Liberty contributing editor G. Patrick Lynch to discuss the price of the pandemic on this episode of the Law & Liberty Podcast.

Related Links

In Covid’s Wake by Stephen Macedo and Frances Lee
Liberalism vs. Scientism, a book review by Mary Carmen Mead
The Pandemic in Hindsight by Law & Liberty‘s Editors (March forum)

Transcript

James Patterson (00:06):

Welcome to the Law & Liberty Podcast. I’m your host, James Patterson. Law & Liberty is an online magazine featuring serious commentary on law, policy, books, and culture, and formed by a commitment to a society of free and responsible people living under the rule of law. Law & Liberty in this podcast are published by Liberty Fund.

G. Patrick Lynch (00:39):

Welcome to the Law & Liberty Podcast. I’m G. Patrick Lynch, a senior fellow at Liberty Fund and a contributing editor to Law & Liberty. And for this episode, I’m filling in for James Patterson. Today we are joined by Stephen Macedo, who is the Lawrence S. Rockefeller Professor of Politics in the University Center for Human Values at Princeton University. He writes and teaches on political theory, ethics, public policy, and law, especially on topics related to liberalism, democracy and citizenship, diversity and civic education, religion and politics, and the family and sexuality. We will be discussing his latest book In Covid’s Wake: How Our Politics Failed Us, co-authored with his Princeton colleague Frances Lee. Macedo and Lee have appeared in many major media outlets, including The New York Times, CNN, PBS NewsHour, and The Hill, to discuss the book, which has been lauded for its critical review of the political and policy responses to the pandemic. Steve, welcome to the Law & Liberty Podcast.

Stephen Macedo (01:36):

Thank you. It’s a pleasure to be here.

G. Patrick Lynch (01:37):

Well, it’s great to have you here. As I mentioned to you before we got started, I think this is a tremendous book and I congratulate you and your co-author, Professor Lee, on it. I want to begin by saying, I know for many of us, looking back on the pandemic is personally difficult because of the challenges we all faced, particularly those of us who lost loved ones. And yet in my view, this book is immensely important because it serves as a sort of a truth commission almost on the Covid response by the government, and it reveals a lot of unpleasant truths about how poorly the public sector performed. The title of your book suggests that the government failed the test of Covid. As a professor, would you have given the government an F for its response on Covid?

Stephen Macedo (02:18):

Well, thank you for the introduction first of all. I haven’t really sat down to think about that. And I think I would say that actually the scope of the focus of the book is a bit wider than just the government, though public health officials and others obviously played a very important role in a way, the focus of the book is on what we call the” truth-seeking departments of liberal democracy,” journalism, science, public health included in that of course, and universities more broadly. And while it’s true that public officials did, I think in many cases pass the buck and say “follow the science” in a way to avoid taking responsibility for hard decisions themselves, it’s also the case that these other institutions did not ask hard questions and didn’t play their role as well as they should have under the influence, I think, of the polarized environment in which we live, a kind of epistemic tribalism that so deeply affects our politics.

(03:20):

So we do look back on the record and try to be as accurate as we can. It’s really only the beginning of a reckoning because there’s so much we don’t understand still. I think about what happened and what the results of the policies may have been, but it’s our best shot at doing a first take, fair mindedly, on what happened. And we do emphasize the shortcomings. Part of it is that there were a bunch of pre-Covid pandemic planning documents, which looked at these non-pharmaceutical measures going back to 2005, 2006, and they were skeptical, let’s put it that way. They emphasized the weak evidentiary basis for a lot of those measures that were imposed. And they emphasized the importance of being frank with the public about the weak evidentiary basis and the certainty of costs, considerable costs from imposing these measures.

G. Patrick Lynch (04:18):

I’m glad you raised that because I think that was one of the most eye-opening parts of the book. I think if you asked the average citizen, if they knew that, in fact, all the pre-pandemic planning had shown that there was skepticism about many of the measures that were pursued, they would be surprised. And I certainly was surprised. Can you explain to our listeners exactly why policy officials in most of these major health organizations in the public health field generally advocated for abandoning most of this planning? What happened? What shifted in their thinking?

Stephen Macedo (04:47):

Well, let me say a word about those things first. It is remarkable that the planning documents went back early to 2005, 2006. George W. Bush read John Barry’s book on the great influenza of 1918, and he initiated these planning processes. There was always a national security component as well, having to do with bioweapons, but he was concerned about another pandemic. There were mathematical modelers who were optimistic about the effectiveness of these measures, very optimistic. And they wound up playing a leading role as it turned out in shaping some of the discourse. But going back to that era, scientific officials who looked at this, including the Institute of Medicine, as it was called back then, which has issued a report on these measures and said that they found the evidence unconvincing and that there was a lot of uncertainty around these measures. And anyway, in 2019, Johns Hopkins did one of these surveys on all of these non-pharmaceutical measures, social distancing, mask-wearing, school closures, border closures, and so on.

(05:49):

And emphasized just as you said, the weak basis. And the World Health Organization did one that came out in November 2019. And among the things that said there, again emphasizing uncertainty, weak evidence, was that there were certain measures that were not recommended under any circumstances, including border closures, quarantine of individuals who had been exposed, and contact tracing, and those were not recommended under any circumstances according to the WHO. And those were all implemented in short order during Covid. I think what happened and what shifted the discourse was partly that the World Health Organization sent a team to China in late January 2020 to survey their response. And that team came back after spending about seven or eight days in China in Wuhan, Hubei Province, and proclaimed that the Chinese had solved the problem, that they had defeated the virus, and that every government in the world should follow the Chinese strategy of suppressing the virus, locking down–something that’s never really been attempted in human history before.

(06:55):

And when you read that report from the World Health Organization, a joint mission that included half Chinese participants, half non-Chinese participants, only a couple of Americans were involved in that, it reads like an award citation, as we say in the book, it’s just uncritically endorses the strategy. So that played a major role. And then there were a couple of other things too. A bombshell report came out of Imperial College London on March 16, which had very impressive graphs. The lead there was Neil M. Ferguson, who’s another mathematical modeler who had been involved in the original 2005, 2006 projections. He said that if we didn’t impose this whole suite of measures, this “Report Nine” from Imperial College said that if we didn’t impose this whole suite of measures by the end of August 2020, we’d have 2.2 million dead Americans. So that report, which was finalized and published on March 16, the Trump administration had gotten an advanced copy of it, that was shown to Donald Trump on March 16.

(07:58):

And later that day, he had a news conference where it was recommended that places with infections spreading should close bars and close other places and impose lockdowns. By that time, schools had already closed across substantial parts of the country. But so those two things were important, and there were other projections that were made. 3.4 percent death rate was projected by the head of the World Health Organization in, I think it was February, or maybe it was early March. So there was a lot of fear. There was a lot of uncertainty of course, and the original lockdowns were very widespread across the country and across Europe. But I think what’s in a way most important is that the European–well, some countries never closed their schools at all. I don’t think Iceland ever closed at schools. Sweden never closed them for kids under 16. And the Western European countries in general, France and Germany and other places in Europe, reopened their schools at the end of April and May.

(09:03):

And it was clear, by the middle of May certainly, that schools were not spreaders of Covid and that teachers were not at great risk. And the European health ministers met education ministers, excuse me, met I think it was around May 10 or something, and issued a statement of the education ministers from around Europe saying that it was safe to reopen schools. So there was a lack of learning, especially in Democratic states, which closed down faster and stayed closed much longer than Republican states. It was a partisan pandemic, as we say in one of our chapters, that there was a striking partisan divide on the stringency of all of these measures with democratic states imposing school closures much longer. Republican states were generally reopening their schools in the fall of 2020. Many Democratic school districts were closed through almost the entire 2021 school year. California closed right through March. So the discourse, the narrative, became very rigid and intolerant of dissent. So I would say that a big part of the book focuses on the shortcomings of public deliberation and kind of intolerance of dissent, forgetting the fundamental principles of science and, in a way, classical liberalism: To be open to dissent, to allow criticism, to welcome criticism, and to not close off disagreement prematurely.

(10:37):

I think one thing that happened, we found is that, about April 2020, once the lockdown strategy had been adopted in, of course, many parts of the country, but there’s the sense of a war mentality that arose. We see this in a report that came out of the Safra Center at Harvard that was endorsed by the Rockefeller Foundation and other foundations including the Niskanen Center, and indeed the American Enterprise Institute put out a Covid planning document or policy statement around that time, April, May 2020, which also endorsed the lockdown strategies to some considerable degree descent, became that this is our chosen strategy and dissent was unwelcome, that we were on a war footing and we needed the equivalent of national unity in the face of a war, in this case, a war on Covid. So as in previous wars, we cracked down on dissent, and that was the big problem. Mistakes get made, but we should have been more open to dissent and disagreement and more ready, prepared to learn over the course of 2020, especially over the summer.

G. Patrick Lynch (11:52):

So this is one of the questions I wanted to ask you, because I think the emphasis on deliberation, discourse, and exchange of ideas is one of the strengths of the book. You mentioned, for example, the Great Barrington Declaration in the book and talk about it in a way that I think is probably the fairest and most honest assessment of it that I’ve seen because it was such a polarizing document when it first came out. And you also mentioned the example of Sweden. Before we get into the sort of who was right question, can we talk a little bit about the importance of deliberation during these times of crises? I guess I was struck by, on the one hand, the hopefulness that y’all place in the importance of having that deliberation discourse during a crisis, during this war language. But is that too much to ask of democracy? Is it too difficult to believe that during a crisis we can tolerate and have that kind of discourse and exchange? And I think this is obviously critical, looking back on Covid, and potentially even more important for looking forward to whatever the next problem’s going to be.

Stephen Macedo (12:53):

Yeah, well, it is a very reasonable question, and I don’t have a particularly good answer for it. One thing I would say is that other democracies did a better job. There was more open debate, I think, in the UK, there’s a very good book by Mark Woolhouse called The Year the World Went Mad. He’s obviously very critical of the lockdown strategies, as you can tell from the title of the book, though he supported the first lockdown, though in retrospect, he doesn’t think it did much good because these lockdowns tend to be imposed after the viruses, the horses out of the barn. So the lockdowns tend to get imposed, late viruses tend to come in waves. So oftentimes the rates of infection will drop at some point after, and people assume that’s because of the interventions, but it just turns out there’s not a lot of great evidence for that.

(13:39):

But they did learn more over the summer in the UK. There was more open debate there. And in Germany and France, I think. So the US is especially polarized by Western standards, and this is a dispiriting picture, a window into the state of our highly polarized political discourse. So I wouldn’t hazard guesses about how well we’ll do in the future, but we do need to think about this. And part of what happened, I think, is, as you’ll recall, 2020 was an election year. Donald Trump was on the ballot. He was associated with the “Let’s reopen the economy as fast as we can” sort of approach. He also, of course, referred to the virus as the “China virus” and so on, kind of unhelpfully actually. But suggesting that the virus might have leaked from the lab was always a plausible hypothesis, which again, should have been entertained and was not particularly, and still not being properly entertained, I think, in the US more so than in Europe.

(14:37):

So I think that was part of it. And of course finally there is some reckoning going on on other issues on this too, that if Donald Trump said it, that must be the opposite of the truth. And that was one thing to say we should not take him particularly seriously on these particular matters, but it’s another thing to assume that the truth is the opposite of what he says. That was a silly approach as well. But deliberation is important. It proceeded better in other places. We should have paid more attention to the pre-Covid pandemic planning documents. And there were voices. I mean, it’s not as though no one pointed these things out. Emily Oster, an economist at Brown, wrote some very good pieces pointing out the cost of school closures and the safety of reopening them. Graham Allison, who is scholar in crisis decision making, going back to the Kennedy administration, he’s a very senior scholar at Harvard.

(15:35):

He put out a statement right around the time that Harvard Saffer Center put out its statement, and this is I think in April, maybe late April 2020. And Alison said, look, we need a wider conversation and wider deliberation here. We need a Team B, we need devil’s advocates. We’re doubling down on the strategy and we’re not sure that it’s the right strategy. It’s going to be costly, he pointed out, and we need to reexamine our assumptions and widen the discussion beyond epidemiologists and public health experts, and have a wider conversation. And he was exactly right, surely on that score, but from what I can tell, his statement was completely ignored. So there were people calling for wider dissent, and there were others. Katherine Eban in Vanity Fair wrote some good pieces as well. So there were a few voices out there, but these issues were not discussed adequately for sure.

(16:35):

And then the Great Barrington Declaration–well, let me just say one other thing. When the lockdowns were imposed in March, there were dissenters. Michael Osterholm, who’s one of the leading epidemiologists scientists in the world at the Mayo Clinic University of Minnesota. He warned in mid-March 2020 that these lockdown measures were unlikely to succeed, that they would be extremely costly, that the best strategy was probably protecting the vulnerable from Covid and allowing most people to go about their business, and recognizing that there was a huge age gradient in vulnerability here. Young people were relatively at low risk. They could acquire herd immunity from getting infected and recovering. He pointed all this out in March. David Katz from Yale did. Tom Frieden, former CDC director, again, pointed out the cost of schools back in March 2020. And then those voices tended to become silent. And when we got to October and the Great Barrington Declaration was published, they were saying essentially the same things–Kulldorff, Bhattacharya, and Sunetra Gupta from Oxford, Harvard, and Stanford, they were essentially saying essentially the same things that were said back in March by Michael Osterholm and Tom Frieden and David Katz at Yale and Justin Leslie from Johns Hopkins. But they were greeted, as people know, with great intolerance. Francis Collins, the head of the NIH then, called them “fringe epidemiologists, dangerous, they’re going to kill people.” By that time, schools had reopened in Europe and reopened for months. Schools had been open for Republican states since August. So I think when it comes to school closures, I think people are starting to recognize that that was kind of egregious, the extent to which we didn’t learn from other places about the relative safety of reopening schools. And so I do think that while the Great Barrington people were wrong about some things for sure, they speculated about the virus having traversed society in a much more widespread way over the summer.

(18:37):

And in the fall, there were some projections about the virus being already on its way out over the summer. I think it was what Sunetra Gupta said, and Bhattacharya has gotten a lot of flack for a Wall Street Journal op-ed he wrote with, I think it was Eran Bendavid, in I think it was April 2020, March or April 2020, suggesting that Covid might lead to as few as 20 to 40, 60,000 deaths. That was the low range of his speculation at that point. The main point of that op-ed and of his work in the spring of 2020 was trying to do studies of the population to see how many people had antibodies or other evidence of having had Covid infections. We knew how many people were getting seriously ill from Covid, but we didn’t know how many people had contracted the virus and not gotten seriously ill because the symptoms are much the same as the flu or cold or other symptoms.

(19:35):

And many people had mild symptoms. So he was really in that op-ed emphasizing the importance of doing these seroprevalence studies in the community to try to figure out the actual mortality rate, which was not as low as he speculated. But there have been viruses in the past that have more or less just disappeared: Swine flu in 1976, we had a mass vaccination program for that, and it never materialized. So he wasn’t completely off base when it comes to the evidence of previous flus. But in any case, his main point was that we needed to do these seroprevalence studies, and he tried to do them in California on a shoe string, but the CDC should have been doing them and wasn’t doing them at the time. So I think the Great Barrington folks were hardly infallible. And Bhattacharya has admitted himself that he made mistakes, everyone did. But I think they were calling for a wider discussion of the cost of these measures. And certainly in retrospect, they were correct about a number of things, including the need for a wider discussion. And going forward, we should at least take seriously respectable dissenters as they certainly were.

G. Patrick Lynch (20:51):

So you mentioned Emily Oster and Graham Allison. It’s interesting. Those were two people I wanted to sort introduce in the conversation because Emily Oster’s work was to try to figure out, she was trying to collect data on schools and just take a look at the trade-offs, and to raise that trade-off question. One of the interesting things for me, reading this boo,k was learning about what worked and what didn’t. You’ve got an entire chapter dedicated to taking a look at the US through the lens of federalism to try to see if a lot of these NPIs were effective. And can you walk our listeners through sort of a general overview of those results so that you can at least share with them what y’all’s view is on the effectiveness of the NPIs compared to vaccines, compared to some of the other measures that were

Stephen Macedo (21:35):

Pursued? Yeah, yeah. My coauthor, Frances Lee, who’s an excellent empirical scholar, very broad intellect, but she really knows her empirics as well. She did a 50-state study of the United States based on CDC data from across the 50 states, looking at Covid mortality from the period from the beginning of the pandemic through the availability of vaccines. And it’s not the best data in the world, it’s state by state, but it’s the best we can do. Other scholars have done this work. Thomas Bollyky also did a state-by-state study using somewhat different data. He came to the same results. But what it shows is that the states divided very sharply in terms of the stringency of these Covid interventions, school closures, masking, social distancing, business closures, restrictions on social gatherings, and so on, with Democratic states imposing much lengthier stay-at-home orders compared to Republican states.

(22:33):

It’s really a very strong partisan relationship, very strikingly so. And there was also a partisan relationship on vaccine uptake with Democratic states people getting vaccinated at a much greater rate than in Republican states. And what the data that we have shows on mortality in the 50 states is that there was no relationship between the imposition of these non-pharmaceutical interventions, the school closures, the masking, the social distancing requirements, stay-at-home orders, and so on. And Oxford has this index of nine of these measures, and they’ve rated governments around the world on the stringency of their interventions, and we have our own data on school closures and so on. No relationship between the stringency or length of these restrictions and Covid mortality. The partisan divergence on mortality begins after vaccines are available. So the evidence shows that the evidence that we have provides evidence for the effectiveness of vaccinations at curbing mortality, the rate of death correcting for age and percentage, uninsured and percentage urban, correcting for a whole bunch of other confounding variables.

(23:46):

The evidence shows that the rate of mortality did go up in Republican states by about a third higher than Democratic states after vaccination, but not before. So we simply have an absence of evidence for the effectiveness of the NPIs on mortality. There is some evidence that they reduce transmission, but not enough or in the right places to affect mortality. We just had a piece published in the Boston Review Forum, and there were four critics there, three of whom focus on this issue, expressing doubt about our findings, but they all emphasized transmission reductions. None of them present any evidence suggesting that there was mortality reduction. And it’s just not clear that these costly measures were worth it if we didn’t reduce mortality and serious illness. Now, there’s another study in Europe, 29 states of Europe by Pizzato et al from the University of Milan, came out in 2024.

(24:43):

It’s excess mortality data across all the states of Europe with a 10 year prior to Covid baseline. So what they did is they looked at the mortality rates across Europe and the 10 years prior to Covid, they adjust for demographic changes over that course of that time and under Covid and look at the excess mortality post-Covid just to see that that’s considered the best measure because when it comes to recording Covid deaths, countries have done it very differently. In Norway, for example, in order to be recorded as a Covid death, it had to be certified by the medical examiner. In Sweden, if you died with Covid, it was counted as a Covid death. So countries had different methodologies for counting their Covid deaths, but if you focus on excess mortality, that’s considered the better study. And what this study by Pizzato et al show, is that Sweden had the best outcomes across the pandemic in the entire 29 states of Europe.

(25:36):

They did have a bit higher earlier on in the first few months. There may have been various reasons for that. They had had a mild flu season the year before, maybe even the two years before as compared with Norway and some other countries which had harder flu seasons. So there were more people alive who were very vulnerable to respiratory infections. But by the spring of 2020, they were doing better than their Nordic neighbors and other European countries. And over the three years of the pandemic, they had the best outcome in Europe according to Pizatto et al. And again, they find no relationship, and they expressed surprise about this and the article, no relationship between the more stringent NPIs and lower mortality across Europe. So the evidence is fairly considerable, and it just means, as my co-author says, and in the language of social science, we cannot reject the null hypothesis, which is we just can’t reject the proposition. The evidence does not allow us to reject the proposition that these non-pharmaceutical interventions, costly as they were not just economically, but in terms of our lives, socially, family relations and so on, social relations. We just can’t show that they had any significant effect. So that’s what the data shows.

G. Patrick Lynch (26:49):

Steve, where did discussions about trade-offs go exactly? Because you guys have a great citation about a survey of economists who were asked whether or not they supported the lockdowns despite all the costs. And something like, I believe the site was something like 80 percent of them said, no, the lockdowns are worth it. And these are economists who were supposed to be talking about trade-offs. And y’all point out throughout the book that there was no consideration or at least very little consideration, except for some dissenting voices to the notion that the costs here were significant, there were things that had to be included in the policy discussion. What happened to alternative voices? What happened to these kinds of discussions? And can you talk a little bit about some of the suppression stories that you tell in this book about the government and certainly leaders in public health and leaders in social media and their inability to maintain the kind of liberal values that you think are very important?

Stephen Macedo (27:40):

Yes. Well, I think the discussion of cost was simply in a way bypassed. There was a lot of rhetoric about, oh, the kids will recover. The kids are fine, the kids will be resilient. We do think there was a class bias issue here. The policies were being made by, and of course, the educated elites who staff elite journalism and universities and science, our members, in effect of what we could call the laptop class. We were able to work at home on our computers, being better off. Our kids were relatively well-positioned to be able to study. I mean, it was hard on them, of course. And I don’t think enough attention was paid to the fact that one third of American workers had to keep working right through the pandemic to keep our electricity on and the internet running and the lights on and the fire department and the police department and food service, food preparation and so on and so forth.

(28:34):

So a third of workers had to keep going to work, their kids had to be taken care of, somehow had to do online schooling, and so on, under much more adverse conditions. So there was definitely a class component to this, I think. So that was part of it for sure. Somehow, the cost issue got bypassed. Fear gripped people, the fear of death, dying a suffocating death from Covid. And it was a bad pandemic. We shouldn’t discount that. It was the worst pandemic in a century. But we do find that among journalists and public officials who–take Andrew Cuomo for example, we quote him in the book, he had those news conferences every day, and he won a special Emmy Award for his news conferences. He was seen as the “Un-Trump” during 2020 under Covid. And he constantly said, “I’ll do everything we can to reduce the spread of the virus, and if everything we do saves just one life, it will be worth it” without giving any attention to the other side of the ledger.

(29:36):

So it was like doing cost-benefit analysis with no attention to the costs at all. And there’s actually have a quote from him in the book saying something like having to stay home and keeping your kids out of school very bad, but not death, domestic violence on the increase, very bad, not death. And he goes through this whole series of things. The only thing that they focused on was reducing deaths from this disease. Francis Collins has said subsequently in a really quite amazing online discussion he had in July, 2023, that the way he put it there was that the public health mindset is to focus only on reducing deaths from disease and not to focus on anything else, including keeping kids out of school for so long that there’ll be lifelong detriments that they never will recover from. And of course, bad health outcomes associated with postponing hospital visits, doctor visits, bankruptcies can have very bad psychological and health impacts for those involved owning their own businesses and so on.

(30:40):

So there were going to be deaths on the other side, but no attention was paid to them. And the news media also played a role. They constantly asked why more wasn’t being done. We have quotations from journalists questioning governors and so on saying, and the President, “Is any attention being given to a national lockdown such as being imposed in the state of New York?” and so on. So there was an attention to this. I would recommend one other book by the way, which came out a bit after ours called An Abundance of Caution by David Zweig, which focuses on school issues. And it has many parallels to our book. It’s a terrific book. We think very highly of it. And Frances and I, and it’s well researched, and he really also shows how biased the media was that the percentage of negative coverage of news concerning Covid was overwhelming in the United States as compared with Europe.

(31:32):

So, for instance, he talks about a specific CDC report that came out in April, I think it was April 8, 2020, a month and a half into the pandemic. And it contained various pieces of information, including that children were at very low risk from Covid, but that African Americans and men were at greater risk than women and non-African Americans. The media reported widely on the negative information about men and African Americans being at greater risk, but they didn’t report the positive information about kids being at very low risk, the bad news predominated. And there was a narrative that took hold. It was a partisan narrative. It had to do with partisanship in part, and people weren’t willing to question the narrative. And we moralized the disagreement, questioning what was considered to be the science, not “following the science” was considered morally reprehensible. Masks, of course, became a very visible symbol of this.

(32:34):

The evidence around masking was extremely weak, and people shifted their positions this promiscuously, as did Dr. Fauci. I mean, he was asked in 2019, a few months before the pandemic with no sign of the Covid on the horizon by an interview, what should I do in the event of another pandemic? Should I wear a mask? Dr. Fauci immediately interrupts the interviewer and says, “No, no, no, avoid the paranoid stuff. Just keep yourself healthy, get a good night’s sleep, don’t drink too much, get some exercise, and so on.” And a few weeks later, he’s for not recommending masks, then recommending masks, and so on. And he still says now that they work, but now he says that it had to be an N95, which was not, of course, what was required at the time. And then there were warring op-eds in The New York Times by my colleague, Zeynep Tufekci,

(33:19):

“The science is clear that masks work.” And Bret Stephens, “No, the science is not clear. Well, I think Bret Stephens was closer to the truth there, and the evidence is still lacking for the effectiveness of masks against aerosolized viruses, such as this was. And I think we knew that quite early on. So yeah, the narrative took hold and it still exists, I think, around some of these measures. We’re having some more open discussion, the school closures, the cost of the school closures for kids, I think that’s just too dramatic to ignore. But the origins of the virus issue remains kind of a mystery. Nature magazine, the public-facing magazine of Nature, the science journal, a very prestigious journal, published an article in February 2025 saying, “There’s more and more evidence to support the natural origins theory of Covid in a Pangolin or a raccoon dog from the market.”

(34:15):

I just don’t think that’s the case. There is, it remains debated among scientists, but there’s not growing evidence for that. Only a month later, the French Academy of Science, or maybe it’s the French Academy of Medicine, voted 97 percent to 3 percent in favor of the proposition that the virus likely emerged from the lab. And there is quite a lot of evidence. I think it’s more likely than not that it emerged from the lab, but we’re not having a serious discussion of that. And it’s not just a matter of historical reckoning. Did the worst virus in a century, was it manipulated through research funded in part, funded adjacently, at least by the US government in a lab that involved collaboration between American scientists and Chinese scientists? I mean, my God, we’re not investigating that. But the most important part of it perhaps, is that the research still goes on, that we’re still doing this gain of function of research on coronaviruses and other viruses, and it remains dangerous.

(35:14):

So that I just find shocking as well that we’ve not had more public engagement on that issue. Where’s 60 Minutes when you need them looking into these matters? So there is a question, what mistakes did we make? What should we have done? We don’t really try to answer that question. We’re not epidemiologists. But what was the discussion like as you asked? Did science function the way it should function? Did research function the way it should function? Did journalism function the way it should function? Were skeptical questions asked to authorities? And unfortunately, we find that these institutions did not function as well as they should have. And the skepticism that is normal on the part of these institutions towards public institutions, public measures, public policies, where social scientists routinely question whether any policy actually works. That was sorely lacking around this one.

(36:16):

And we still are surprised. So you asked about restrictions on speech and so on. It did become clear. The evidence presented in the Missouri v. Biden case which became Murthy v. Missouri on appeal, that the Biden administration was suppressing social media posts against US policy during Covid, ample evidence of that, threatening the social media companies with the loss of Section 230 protections against liability, which would’ve bankrupt the companies and put them out of business. And intense pressure was brought to bear because we have the emails now, it was subpoenaed through the records. And this was all being done in secret, suppressing arguments and evidence against government policy and boosting government messaging. And we don’t know actually of a single law school conference on those issues. I don’t think the ACLU filed a brief in that case. So it’s kind of amazing that these matters are not being adequately looked into even now.

(37:18):

And there is concern about misinformation and disinformation on the internet for sure. We’ve always had misinformation and disinformation. Maybe it’s worse now than it used to be, legitimate reason for discussion. But I don’t find myself feeling highly confident about trusting the government to become the censor here. So that record as well is quite important and quite impressive. We talked to a scholar who also edits an education journal, and he wrote a piece in March or April 2020, pointing out the cost of school closures, drawing out evidence from previous teacher strikes and blizzards where schools were closed even for just a couple of weeks or a few weeks, and pointing out based on that evidence, how much learning loss would take place and so on. And he was saying to us that as far as he could know, that that was immediately suppressed. And his journal more or less disappeared from the internet once he published that article. So that’s anecdotal, but we have ample evidence that that sort of thing was happening partly on the initiative of the social media companies themselves, but then certainly with administration involvement under President Biden, which they renounced it. And so when it got to the Supreme Court, the Supreme Court didn’t consider merits because the policy had been revoked, but it certainly is a matter of concern. Free speech was one of the casualties of Covid it seems, or at least unbiased representation of different points of view on the internet.

G. Patrick Lynch (38:52):

So two quick questions to finish up. Number one, you talked about the chapter in which you all summarized the cost. Can you briefly walk our listeners through some of the costs of the pandemic as you all measure it?

Stephen Macedo (39:02):

Yeah, this will be brief. We have a whole chapter on that. And again, my colleague took the lead on that and did a wonderful job. Well, there are the fiscal costs for one thing, the 2020 Covid relief effort was as much as a percentage of GDP corrected historically as the entire New Deal and the 2009 financial rescue package put together. So 2020 was a huge fiscal cost. We have graphs there showing that. And then in 2021, we did essentially the whole New Deal over again as a percentage of GDP, as the equivalent of percentage of GDP. So the fiscal cost and the debt burden that we passed on to future generations is itself quite significant. And it also means that the country is in a weaker position when it comes to trying to respond in a fiscal way to a future recession.

(39:54):

Learning losses have been profound. I believe it’s the case that chronic absenteeism is up one-third from before the pandemic. Losses of learning test scores and so on have been very considerable and have been especially bad among poorer kids, low-income kids, and minority kids, just as predicted. So the learning losses and the lifelong detriments to education are profound. Associated with the loss of jobs and schooling, the crime rate went up in many cities, the biggest year-on-year increase in crime in recorded history, I believe. At first, that was associated with the George Floyd protests. But recent research now shows that they pre-date that the crime surge is associated with the Covid lockdown measures and business closures, not the George Floyd protest. They may have contributed a bit later on. Again, business losses and healthcare losses in terms of postponed doctor’s visits, medical interventions, cancer screenings, and so on, depression, higher rates of depression, drug abuse, and so on, including the populations, not themselves significantly affected by Covid. And greater isolation. Church attendance has, I believe, never returned to pre-Covid levels, business closures in many cities.

(41:20):

If you go to Washington, DC, the city is still never fully recovered from Covid, nor is San Francisco. So there have been many losses of businesses in inner city areas that have permanently, or at least over the long term, reshaped many cities, and the feelings of depression and isolation and loss of social connection, and so on. So the cost themselves have been quite considerable, and we’re still trying to figure them out. I think our chapter is a kind of first take survey on that, but there’s no question that the costs are very considerable, and for many people will be lifelong. And just as the pre-Covid pandemic plan said, they seem to have hit poor and minority families and children, and people more severely than the better off.

G. Patrick Lynch (42:10):

I want to finish with one final question. One of the big takeaways for me of this book was that humility is something that we could use a lot more of, certainly in public life, but also in politics. Because I was stunned at how policymakers and politicians were either afraid to say or didn’t feel comfortable saying, We don’t know. We don’t know exactly what the trade-offs are. We don’t know exactly what works and what doesn’t work. We’re going to try things. We’re going to try to move forward with them. And I guess I wonder a little bit, a big part of the project that you all are pursuing is this focus on democratic institutions and elites and other policymakers and folks in academia and media, their inability to accept humility. Is an ability to accept humility one of the goals of this book? Is it something you guys thought about and talked about as you were composing it? Because it seems to me my big takeaway is we need to be a lot more humble the next time we face something like this to understand what we don’t know before we start acting on the things that we think we know.

Stephen Macedo (43:13):

I think that’s exactly right. I think you’ve captured it exactly. There’s no question that those pre-Covid pandemic plans that I mentioned, which are quite remarkable to go back and read, and we survey them in our chapter…

G. Patrick Lynch (43:24):

That was an eye-opening way to start the book.

Stephen Macedo (43:26):

It’s just astonishing. I was sort of amazed when I came across them and even this Institute of Medicine letter from 2005 warning that public officials will be apt to exaggerate the certainty of these measures working and to implement them to show that they were in charge to show that they had things under control for political reasons, in other words, not because of the sound science behind them. So I think you’re exactly right. And I think part of this is a public health thing that the messaging from public health officials and public officials associated with pandemic planning probably had to do with behavioral change, trying to get people to change their behavior. Francis Collins says this, we quote this in the book that we were trying to get people to change their behavior, in case what we were recommending was right, but they weren’t sure it was right.

(44:16):

And he admits again in the July 2023 message he gave, that we should have been more frank about that. And we lost a lot of trust because we were not frank, that we weren’t sure about the things that we were recommending. So I think that’s a big part of it, is that public health officials need to realize that they are responsible for telling the public the truth. They have not been authorized to not share the truth with us. And the way that Paul Offit, who’s a doctor at the Children’s Hospital of Pennsylvania, puts it, is they need to commit themselves to truth-telling with nuance. If there’s uncertainty communicated, if there is some evidence of adverse incidents associated with a vaccine, even if they’re quite minor, don’t cover it up, allow it, and recognize the limits of the evidence. And that’s going to be the only way to restore trust is to be frank with the public, because the public has access to many sources of information and many dissenting voices.

(45:08):

And I think honesty and humility are two virtues that need to be rediscovered in the wake of Covid. We just need to be honest about the limits of the evidence. People figure these things out. They were told about the vaccination. So if you get vaccinated, you’ll be a dead end to the virus. You will not transmit the virus to others was what was being suggested. We did not have evidence of that. Transmissibility was not an endpoint in the vaccine trials, so that should not have been claimed. And once people found that, well, even if they got vaccinated, they could still get Covid, get milder, somewhat severe disease, but not life-threatening disease nearly at the same rates. Nevertheless, they felt they had been misled. So I think you’re exactly right. We need to be much clearer about the limits of our knowledge. It’s not a comfortable thing to hear in a pandemic that we don’t know exactly what to do.

(45:58):

We’re not sure, or maybe this is worth trying even though it’s going to be costly and there’s some evidence for it. But I think that that’s absolutely necessary: frankness and truth-telling. And those are supposed to be, again, the fundamental virtues of science and universities, university researchers, and journalism. So I agree. I think that that’s exactly right. And that’s something we found throughout, that there was an inadequate commitment to frankness, humility, and also being willing to say things that might seem to be critical of your own side in a partisan way, things that might be at odds with the narrative that’s favored by your side and the partisan struggle. I think people need to rise above. Scientists, journalists, academics need to do more to realize the importance of rising above partisanship.

G. Patrick Lynch (46:55):

And on that hopefully positive note, we will conclude. Steve, I’d like to thank you very much for being on the Law & Liberty Podcast.

Stephen Macedo (47:01):

Today. Thank you very much. I enjoyed the conversation.

James Patterson (47:05):

Thanks for listening to this episode of Law & Liberty Podcast. Be sure to subscribe on Apple, Spotify, or wherever you get your podcasts. And visit us online at www.lawliberty.org.

Freedom of speech is in a bad place. The modern, young identitarian left appears to have given up on it, regarding it as a dinosaur, of value only to bigots, billionaires, and kleptocrats. Elements of the political right regard it as little more than a political football, to be kicked about in the culture wars as a weapon in the war on woke. The former are more interested in cancelling speakers who offend them than in engaging them in debate and seeking to defeat them in argument. The latter take an extremist approach to free speech (as long as it is speech they agree with), holding that anything goes in an unlicensed, unregulated marketplace of ideas.

Both are profoundly wrong and are doing immense damage to the once-cherished ideal of free speech. To help rescue free speech from the cesspit of the culture wars, there are two places we can look: to the past and to the world elsewhere. If free speech is in the deepest trouble in the United States, is it quite as bad elsewhere in the Anglosphere and is it in a better place beyond the English-speaking world? And if free speech is in trouble now, how did we get here? Was it always thus, or was a wrong turn taken at some point on the West’s journey from the eras of heresy and treason, sedition and blasphemy (all of which were crimes capable of being committed by words alone)? These are among the questions Fara Dabhoiwala asks in his new book, What is Free Speech? A historian now at Princeton who has formerly worked in Oxford and elsewhere, Dabhoiwala is well placed to address them.

What Dabhoiwala gets right is his framing: the way he sets up the problem is excellent, and the book’s introduction and opening two chapters are its strongest sections. He is right that free speech “has always been a weaponized mantra,” that understanding free speech “is also, always, about uncovering the unequal distribution of power,” that free speech has often been regarded as dangerous not only for its power to “unsettle orthodoxy” but also by reason of its being “perpetually manipulated by the powerful.” He is right that “wherever one looks, liberty of speech was never a stable concept.” He is right that, in the history of human development, it is a recent idea and that for most of recorded history it was simply “unthinkable”—not even “intelligible” as a concept. He is also right that, in the English-speaking world, Trenchard and Gordon’s Cato’s Letters (1721–23) and John Stuart Mill’s On Liberty (1859) did more than any other works to articulate, celebrate, and popularize notions of free speech.

Or, at least, he is very nearly right about that. To these twin canonical texts, I would add a third, John Milton’s Areopagitica (1644), the importance of which Dabhoiwala downplays. Milton’s great essay, for all its partisanship and other flaws, remains the most powerful tract in our language against pre-publication censorship. It is curious that Dabhoiwala should undervalue it, for Milton’s argument actually accords with the conclusion in What is Free Speech?, where the interests of the audience are privileged as much as the interests of the speaker. Milton was fundamentally opposed to censorship, not because it diminishes the rights of the author (or printer or publisher), but because it infantilizes the rights of all of us as readers to choose for ourselves what we should read. Dabhoiwala would agree with that, and his argument in the closing section of his book would have been all the stronger had he harnessed Milton in its support.

Cato’s Letters and On Liberty are far from perfect accounts of free speech. There is a marked naivete about Trenchard and Gordon’s blithe assumption that the truth will always and inevitably emerge if only “all opinions are equally indulged and all parties equally allowed to speak their minds.” And there is a notorious silence—or gap—in Mill’s “harm principle” owing to his refusal to define what a harm is when it comes to what he called “the liberty of thought and discussion.” Is hate speech harmful? Is offensiveness a harm that can justify the curtailment of speech? Is misinformation? Mill answers none of these questions. Dabhoiwala devotes much of his book to challenging Cato’s and Mill’s arguments for free speech.

Free speech does not empower us all equally, but its absence empowers only the censors, whatever garb they clothe themselves in.

Neither Cato’s Letters nor On Liberty should be regarded as sacred texts. Neither is above criticism. Some of Dabhoiwala’s critique is astutely judged, particularly of Trenchard and Gordon’s failure to see how cut-throat competition amongst newspapers and problems of media partisanship undermine several of their arguments for a free press. But Dabhoiwala sets out not to rescue Cato and Mill from their mistakes and oversights, but instead to trash them. That approach is as baffling as it is bizarre.

Trenchard and Gordon were writing in the 1720s avowedly about political speech. In that era, the polis was composed entirely of men. Cato has nothing to say about women’s right to free speech. Does that undermine any aspect of Trenchard and Gordon’s arguments against the official abuse of laws of sedition and libel? Mill worked for the East India Company. His focus in On Liberty is on the liberty of thought and discussion in England, not in India. Does that undermine any aspect of its argument? Dabhoiwala seems to think so. Yet, in whose interests is it to draw our attention to Mill’s early work as a civil servant in the East India Company but not to his pioneering campaign later in life as a Member of Parliament in striving to bring Eyre to justice for the racist atrocities he committed as Governor in Jamaica? All his life, Mill campaigned for women’s rights. Why fail to mention this when so much scorn has been poured on Trenchard and Gordon’s failure to accommodate women’s free speech in their account? We all know what gesture politics is. There is such a thing as gesture history, too. Much of Dabhoiwala’s account reads more as a history against free speech than as a history of free speech.

Speech and Truth

The reason for this is that Dabhoiwala appears to crave not free speech as much as true speech. Throughout his accounts of Cato and Mill, he is captivated above all by their assertions about truth. He is right that what they have to say about truth can be unpersuasive. Free speech has not acted as a guarantor of truth, and Cato was wrong to suggest that it would. Mill, however, did not repeat that mistake. His argument was that free speech was needed in order to enable us to seek the truth, not that free speech would necessarily ensure that the true would drown out the false. (In this regard, Mill’s position was closer to Milton’s than it was to Cato’s.) Mill did not regard any argument from truth as his most important contention for free speech. At the core of his vision for liberty of thought and discussion lay his notion of individual human flourishing. We can grow and develop, and our judgement can be valued, Mill argued, only when we are exposed to that which we do not already know—to new facts, to divergent opinions, and to novel interpretations. That is not only a much stronger argument for free speech than any that relies on truth, but it is also an argument for genuinely free speech and not an argument only for true speech.

The best argument for free speech, however, is to imagine a world without it. We do not have to imagine very hard. Radicals and revolutionaries of many stripes have been so convinced that they are right and everyone else wrong that they have readily grabbed and grasped at the levers of control over discussion. Robespierre did it. So too did Lenin. Neither was very concerned with freedom. Both were very concerned with what they thought was true. And, as Dabhoiwala observes, we know how that story ends. The privileging of true speech over free speech has been tried. It was tried by the medieval church in the age of heresy. It was tried by pre-Enlightenment governments in the age of sedition. It was tried by the totalitarian regime of the socialist Soviet Union. It has been tried and, repeatedly, it has failed.

One of the greatest strengths of Dabhoiwala’s book is his insistence that free speech regimes distribute power unequally. Media moguls, press barons, and the owners of social media platforms wield immense power over their viewers, readers, and users. But when we imagine (or recall) a world without free speech, we conjure a world where the church, the state, or the vigilante goons who wish only to cancel and destroy that with which they disagree run riot. Free speech does not empower us all equally, but its absence empowers only the censors, whatever garb they clothe themselves in, whether they be priests, inquisitors, ministers, or culture warriors.

Free speech needs to be rescued from the culture wars, not condemned, its pioneers pilloried for having failed to prevent them.

What then must we do? If America is drowning in lies, if social media is awash with torrents of misinformation, how might we work our way out of the abyss without giving up on the great liberal values of freedom and autonomy in thought and discussion? There is an answer. In a word, it is balance. Free speech is of immense value. But it is only one value in a pantheon of principles. Dignity is also of value, as are accuracy, honesty, impartiality, decency, personal privacy, and a whole constellation of matters which may sometimes clash and come into conflict with freedom of speech. Social life, political life, and the pursuit of happiness depend on the liberty of speech being balanced with other public goods such as these. Everywhere this is understood, even in the United States.

American Exceptionalism

It is often claimed, including by Dabhoiwala, that American exceptionalism privileges free speech as an absolute, as a card that trumps all competing values and principles. Yet this is not quite so. It is true that in the modern United States, aspects of free speech are more heavily protected than is the case in Canada, Britain, or Europe. But this is not true for all speech and even in the US, there are some forms of speech that are not legally protected at all, such as obscenity, child pornography, defamatory speech, and fighting words. Moreover, the over-protection of speech is a recent development in American jurisprudence—it dates only from the 1960s. Many plausible interpretations of the First Amendment are available, which are far less militant—far more balanced—than the one which is usually preferred by a majority of today’s Supreme Court justices. Care should be taken, though, not to exaggerate the extent of the differences between American and (say) British or European jurisprudence on free speech. For sure, there are legislative restrictions on speech which are nowadays ruled unconstitutional in the US, which would be lawful in Britain or Europe—restrictions on campaign expenditure or on hate speech, for example. But there are also many more cases where a balance is struck between speech and some countervailing interest, such as security (think of the recent TikTok case) or dignity (think of Virginia v. Black, upholding the criminalization of cross-burning). Free speech is a powerful right in US constitutional law, and there are instances, particularly in the last twenty or so years, of it being militantly enforced, but even in America, even today, it is very far from being an absolute.

Once we accept that free speech is a value which must be balanced against others (as Dabhoiwala does and as I would agree), we have to start thinking hard about who should undertake the balancing exercise and how they might achieve it. This is precisely the difficult, often laboriously detailed work that Dabhoiwala acknowledges needs to be undertaken. Yet he disdains from doing it himself. Others, thankfully, are working on it. European human rights law, for example, has developed over more than fifty years a sophisticated account of how freedom of expression must be balanced in a proportionate manner against other competing rights and interests. It is a pity Dabhoiwala has nothing to say about this.

Or take another example. Britain’s Online Safety Act, passed by the UK Parliament in 2023, is one of the world’s leading legislative attempts to bring a degree of regulation to the Internet—particularly to search engines and to social media platforms—with a view to making it a safer space, especially for children. The legislation runs to hundreds of pages. It took over four and a half years to enact. It will be enforced by a statutory regulator, Ofcom, which has already published thousands of pages of guidance and consultation documentation on the Act’s implementation. It is an enormous enterprise. It makes an avowed attempt to balance freedom of online expression against the harms that speech can cause, particularly speech with the speed and reach of which social media boasts. It is both vital and painstaking work. Yet it merits a mere seven and a half lines in Dabhoiwala’s more than 400-page book. In the European Union, a Digital Services Act is doing similar work to the UK’s Online Safety Act. Dabhoiwala gives that legislation a bare paragraph.

These are striking omissions, not least because the ways in which European human rights law and British and European online safety legislation seek to balance free speech against other values are deeply contested and highly controversial. My point is not that they necessarily get the balance right, but that they are, at the least, well-meaning attempts to achieve in practice what Dabhoiwala rightly says needs to be striven for, which is why it is so mystifying that he seems so uninterested in them.

If you want to get serious about how free speech is going to operate in the online ecosystem, if you want to understand how this once beautiful but apparently now merely “dangerous idea” can help to guide us through the myriad difficulties and challenges of Internet regulation, if you are worried about how we should respond to the bilge of misinformation online, you are asking the right questions about free speech. But if you seek insights into what the answers might be, What is Free Speech? is not going to help you. Free speech needs to be rescued from the culture wars, not condemned, its pioneers pilloried for having failed to prevent them.