← Return to search results
Back to Prindle Institute

Rawls

Posted on

Can Algorithms Really Treat Us Fairly?

The world is increasingly inundated by algorithmic decision-making. Everything we experience online is the result of a calculation. It is sometimes difficult to remember that what we see is not chosen randomly, or without reason. There are motivations behind the online experiences we have, mostly driven by engagement metrics and revenue generation.

Beyond our consumption choices, algorithms are used in high stakes, real world conditions. Courts use algorithmic systems like COMPAS to determine the possibility of recidivism. Car insurance companies use AI algorithms to determine whether we should pay more or less money for our insurance. Health care companies are using AI to approve and deny patient coverage.

As our society increasingly embeds algorithmic decision-making into its infrastructure, it will become ever more important to know whether, and precisely how, algorithms can treat us fairly. A great deal of the current research on algorithmic fairness focuses on the best ways to remove biased data inputs and how to adjust outcomes so that they conform to a fair distribution based on some theory of outcome fairness (which is itself a contentious subject). There are, however, far fewer arguments that consider the possibility that algorithmic fairness may be an impossible union, and that some degree of unfairness may always be baked into algorithms. But on what grounds could we make this claim?

The reason is not simply that we can’t make outcomes fair, or that we can’t remove bias from data (though perhaps these things are true). The reason has to do with the very nature of how algorithms work.

To make this argument, I will consider fairness from a Rawlsian perspective. Rawls explicitly and repeatedly states that his conception of fairness concerns the basic structure of society. According to Rawls, the basic structure consists of a political constitution, an independent judiciary, legally recognized forms of property, an economic structure, and a family structure. For Rawls, then, fairness is relevant with respect to these general social practices, including decision making that affects these areas.

Fairness for Rawls also requires engagement between a particular kind of persons, namely rational and reasonable ones. Rational people are those who pursue what is in their own best interests, while reasonable people are those who are willing both to propose fair terms of cooperation and to accept them when proposed by others. If we are to be engaged in the creation of a fair social arrangement, we must be engaged with people who are both rational and reasonable, otherwise fairness is likely impossible to create.

Crucially, we must also see one another as free and equal. What makes us equal, for Rawls, is that we possess two basic moral powers: the capacity to form a conception of justice and the capacity to form a conception of the good. What makes us free, according to Rawls, is that we conceive of ourselves and of each other as having the moral power to form, revise, and pursue a conception of the good. We regard ourselves as self-authenticating sources of valid claims. Simply put, we think of ourselves and one another as free.

If we are going to make algorithms fair in a Rawlsian sense, we need to reconcile how algorithms treat people with Rawls’s idea that a fair society is one constituted by free and equal persons. So, what is the argument for why algorithms may be unable to treat us fairly? It goes something like this:

Justice requires treating people as free. This entails (1) treating people as capable of conceiving of the good and pursuing it, and (2) treating people as self-authenticating sources of valid claims.

However, algorithms cannot, in principle, do this. Algorithmic systems do not have minds, and so they do not think of us as free, for the simple reason that they do not think at all. Treating someone as algorithmically predictable is conceptually incompatible with treating them as free in the Rawlsian sense. Numbers cannot adequately represent the kind of Rawlsian freedom that justice requires.

Algorithms can be deterministic, probabilistic, or non-deterministic. For those who believe that determinism and free will are incompatible, it should be obvious why deterministic algorithms cannot treat people as free. But even for those who accept that determinism and freedom are compatible, a deterministic algorithm still treats humans as deterministic systems, not as free agents.

Probabilistic and non-deterministic algorithms may sound as though they leave room for treating people as free, but it is not clear how they do so. The burden of proof seems to lie with those who claim that such algorithms are fair. If algorithms are supposed to treat people as free, then we need a reason to believe that treating people probabilistically or non-deterministically is equivalent to treating them as free.

Consider a concrete example. A court uses COMPAS to evaluate a defendant’s eligibility for bail. After using the probabilistic algorithm, the court decides to deny bail. The algorithm determines that there is a 75% chance that the defendant, if released, would flee. In what sense does the algorithm treat the defendant as free merely because it relies on a probabilistic calculation?

The crucial question that proponents of algorithmic fairness must answer is this: How can an algorithm treat a human being as free when its basic functioning relies on the predictability and quantification of human behavior?

While algorithmic harm and bias are real, there are reasons to be optimistic that we might make algorithms fairer, or more just, on the outcome side. One can define fair outcomes and then adjust decision-making parameters until those outcomes are achieved.

However, if justice and fairness require that our procedures themselves be fair, and if fairness demands treating people as free, then we must also ask whether algorithmic systems can ever satisfy this requirement.

We might wonder why algorithms need to be just in this particular way. One might argue that the solution is straightforward: all that is required is to ensure that a human being remains in the decision-making loop. If a human uses an algorithm and then offers a justification to another human, is that not enough to satisfy the requirement of treating people as free?

I think the answer depends on a few things. First, when a decision-maker justifies a decision by appealing to an algorithm, we can ask whether there are additional reasons for the decision and how central the algorithm’s role was in arriving at it. In cases where the sole justification is the algorithmic output, the presence of a human decision-maker is merely performative. We might as well remove the human entirely and allow an anthropomorphized system to deliver the decision.

However, if there are independent reasons a person can give beyond the algorithmic output, then perhaps the conditions of Rawlsian fairness can be preserved. Even then, this would depend on whether the reasons offered still treat the person subject to the decision as free.

The upshot of this argument is not simply that algorithmic systems can be poorly designed and need to use better, less biased data, or produce fairer outcomes (though surely both would be ideal). Rather, it is that there may be a fundamental mismatch between predictability and justice. If treating people as free requires engaging them as self-authenticating sources of claims, then any system that governs by prediction rather than justification may fall short of fairness in a Rawlsian sense.

This leaves us with an urgent question to answer: are willing to trade procedural justice for the benefits of algorithmic decision-making, or should certain decisions demand forms of reasoning that algorithms cannot, in principle, provide and thus forgo the use of algorithms and AI entirely?

Posted on

Is Civility Overrated?

Marjorie Taylor Greene, once one of the most ardent Trump supporters in Congress, announced her resignation after breaking with the president over his perceived interference with the Epstein files and abandonment of an America first agenda. However, her resignation reflected not merely their political disagreement, but the intense animosity sent her way by Trump, such as branding her a traitor. (Critics were quick to point to Greene’s own participation in divisive political rhetoric.) Famously vituperative, many see Trump’s embrace of vulgarity as part of his initial appeal, which contrasted refreshingly with the cloying civility and polish of an elitist establishment politics. But perhaps civility is worth defending.

An initial challenge is pinning down exactly what is meant by “civility.” In common parlance, civility is entangled with notions of manners, politeness, and etiquette. Philosophers generally want a little more. For example, as discussed by Laura Siscoe in the Prindle Post, the political theorist John Rawl’s duty of civility consists of appealing to shared reasons and presenting arguments and evidence to make the case for one’s political claims. His is not the only account, but it represents a common strand in what philosophers generally want civility to do, namely, facilitate governance. All societies need solutions to the reality of political disagreement. Voting is part of that solution, but civility can be another. For many political theorists, in an ideal political situation, political decision-making would be based on evidence, reasoning, and a shared understanding of where people are coming from, even if they disagree. People present their case and the basis for it, other people respectfully listen, and then votes decide what happens next. Ideally, adopting certain norms and practices of civility can help to ensure that a range of voices are heard, and that decisions are made on sensible bases.

Ideally.

In practice, norms of civility are not usually set by political philosophers, but by established powers in a society. Civility can then become a set of rules for a game played primarily by the powerful. Political activity that does not conform to established standards of civility can be dismissed, not because of its substantial content, but because of form. For example, early advocates for women’s suffrage in the US were often dismissed as mannish, uppity, or lacking decorum. Even with modern protests, debate often centers on whether the protest was done correctly, as opposed to the political issue the protestors themselves care about.

This is not to dispute that political activism and protest can never be done unethically. Political violence especially requires careful justification, as both Richard Gibson and Aaron Schulz have discussed here in the Post. But norms of civility can absolutely be weaponized to quash political action, especially by smaller, poorer, or otherwise less advantaged political communities.

Civility as a norm can also paper over atrocities or the appropriate emotional response to certain activities. Civility welcomes someone into a shared space of reasons and decorum. But is that always appropriate? Could someone’s actions, whether personal or political, merit exclusion from such a shared space, and could failing to treat them with appropriate derision signal that their actions are in fact okay? For example, in 2022 some criticized Joe Biden’s casual friendliness with Saudi Crown Prince Mohammed Bin Salman amid human rights concerns. In short, civility can whitewash crimes and atrocities.

The polished surface of political civility, with its status quo bias, can also hide serious ongoing harms. At its worst, it becomes more problematic whether an action breaks decorum, than whether there is a meaningful ethical concern. We fret about the civility of name-calling and trollish tweets, but rarely about the civility of drone strikes and private prisons. As in the case of violent abolitionists in the 1800s, many people took for granted the familiar violence of slavery, and stood aghast at the resistance of victims.

These concerns are not without counterargument. It might be alleged that the above concerns are about weaponization or abuse of norms of civility, as opposed to civility as such. Or that they really relate to etiquette. Nonetheless, they still represent a practical concern. At a minimum, worthwhile civility requires work, and not just the work of being civil, but ensuring the implementation of civility is actually politically healthy.

Of particular irony is that in these problem cases, civility emerges as potentially harmful to the very democratic values that motivated valuing civility in the first place. It can be captured by the powerful. It prioritizes form over evidence and reason, It often privileges the status quo. It prevents certain voices from impacting the political process. Or at least, these are risks.

Civility also has its limits. For example, if a group has been locked out of politics by the constraints of decorum (or force), then uncivil action of some kind may be justifiable. Just what is justified depends on the exact circumstances, including the degree of current harms, the approaches that have already been tried, and the specific proposed “uncivil” actions. Is it name-calling, property damage, or murder?

What about the particular case of Trump and Greene? Given that Trump is sitting president, it’s challenging to argue that Trump is being excluded from the political process by civility. Nor does he appear to be calling out a specific unacknowledged atrocity. Finally, the effect seems to partly be that she is resigning, and thus removing herself from the political process — an anti-democratic impact. None of this speaks to incivility being of much value in this instance.

It might be objected that on the scale of uncivil action, a few mean truth social posts are hardly worth mentioning. But it gets complicated with public figures with large numbers of followers or supporters. Stochastic terrorism refers to a statistically increased risk of violence when hostile political rhetoric is blended with mass communication. The chance of any given person doing something violent based on an inflammatory tweet is low, but when someone has millions and millions of supporters, it adds up quickly. For such high-profile individuals, incivility is far more directly connected with violence and threats of violence. Consequently, it may be reasonable to expect them to act in a way that minimizes the harm of their rhetoric, to ensure that reason rather than fear of violence pulls the strings of government.

Posted on

Do We Have a Duty of Civility?

black-and-white photograph of gentlemen engaged in gun duel

Data shows that American political culture has experienced a steady decline in civility. This downward shift has become a frequent talking point amongst sociologists, political pundits, as well as many average Americans. Most who work on this issue agree this trend is deeply troubling for the trajectory of American society. So, how can we reinfuse our shared political culture with the virtue of civility?

In political philosophy there is a concept often referred to as the duty of civility. This is a moral (as opposed to a legal) duty to take the beliefs of others in your political community into account when deciding how to vote, which policies to advocate for, and which claims you appeal to in the public square. The origin of the duty of civility is linked to 20th-century political philosopher John Rawls, and his formulation of the principle can be found in the text Political Liberalism. In constructing his political philosophy, Rawls was sensitive to the fact that diverse people living in a free society will inevitably hold disparate opinions and beliefs. The challenge, then, becomes figuring out how to show civility to our neighbors, with whom we might disagree sharply over any number of important matters.

To get slightly more technical, the duty of civility specifically claims that people must be able to appeal to “shared” or “public” reasons as justifications for their political stances. These reasons, according to Rawls, are those your fellow citizens might agree with regardless of their particular religious, ethical, or political beliefs. Rawls took these kinds of reasons to be based in the core values of political liberalism, such as the protection of basic human rights, freedom of conscience, and the protection of one’s political autonomy. The thought here is that all reasonable citizens living in a free and liberal society should be able to comfortably endorse these commitments. Thus, if political policies are always grounded in these shared, public reasons, they have legitimate authority over people, independent of the diversity of opinions that inevitably exists within a political community.

Regardless of whether people agree with Rawls that fulfilling the duty of civility demands appealing to shared or public reasons, the general spirit of that duty seems highly relevant to reversing the troubling trend we are currently witnessing in American political culture. One point of academic debate, however, involves what this duty means for citizens living in non-ideal political communities (i.e., our political community). There is disagreement over whether or not our duty to consider the different, and oftentimes opposing, beliefs of our neighbors is contingent on their willingness to do the same. Is the duty of civility only binding if we can expect that compassion, generosity, empathy to be reciprocated?

Certain moral duties do seem to function in this kind of contingent way. For example, consider the moral obligations generated when entering into contractual agreements with others. Let’s imagine that you enter into a contract with your landlord that outlines the duties of property care they incur in exchange for your monthly rent payment. Let’s then also imagine that they fail to uphold these duties in a significant way. Under such conditions, you are potentially released from your duty to continue paying them. The lack of reciprocity entails the suspension of what would otherwise be your moral duty. On the other hand, other kinds of moral duties seem to persist regardless of whether or not the other party involved holds up their end of the deal. For example, if you have a moral duty to give a certain percentage of your income to charity, this duty does not seem contingent on whether others in your society also choose to donate their money.

I believe the duty of civility is more like the latter example than the former. If a desirable goal for our society is to establish a healthier political culture, it is difficult to imagine how this will occur without individuals taking this kind of duty upon themselves. Forcing ourselves to grapple with the reality that many of our well-intentioned neighbors view the world through a substantially different lens will likely shape the way we engage politics for the better. Embracing the duty of civility helps to prevent the steamrolling of the beliefs, opinions, and convictions of others, which inevitably has positive downstream implications for the health and vitality of our public square.

Of course this is not to say that individuals must bracket off all of their particularly controversial beliefs when it comes to public political engagement. Rather, it is to say that thinking through the implications that various policies you advocate for will have not only on you, but also on your neighbors, is an essential exercise in which to engage. If individuals refuse to engage in such a practice unless enough other members of society also agree to participate, it seems unlikely that the duty of civility will ever become enmeshed in our political culture.

So to return to our original question of how to reinfuse our political culture with civility, it seems clear enough that individuals must strive to fulfill something like the duty of civility, and do so regardless if others around them choose the same. At risk of the further degradation of our public square, it seems incumbent on people to strive for morally ideal action, even in the midst of non-ideal conditions.

Posted on

The Call for Supreme Court Reform

photograph of Supreme Court building facade

Representative Alexandria Ocasio-Cortez has called for reforms to the Supreme Court of the United States (SCOTUS), specifically increasing the number of justices. The reforms are needed, she says, because of the rightward tilt of decisions over the past year including Dobbs v. Jackson Women’s Health Organization (abortion) and 303 Creative LLC v. Elenis (public accommodations and discrimination). On CNN she claimed that if reforms do not occur, then SCOTUS will function “without any check on their power, without any balance on their power, then we will start to see an undemocratic and, frankly, dangerous authoritarian expansion of power in the Supreme Court.” The opinion pages of major journalistic outlets like The Washington Post include similar views decrying judicial behavior.

To assess the need for reform and effectiveness of potential solutions, we must first determine what it would mean to have a properly functioning judicial system. One way to evaluate our social and political institutions is through John Rawls’s constructivist lens presented in his seminal work A Theory of Justice (and elsewhere). Rather than starting with a specific (and contentious) ideal, we tie our conception of proper functioning to the product of particular processes. As Carla Bagnoli describes it, constructivism is a view that “insofar as there are normative truths, they are not fixed by normative facts that are independent of what rational agents would agree to under some specified conditions of choice.”

To make sense of this description I like to think about where constructivism fits into a picture of objectivity and subjectivity. At one end we have facts — statements such as “the cat is on the mat” — whose correctness is determined by a set of referents independent of any human observer. As long as there is a mat, there is a cat, and the two objects have the right kind of relationship, the statement is either correct or incorrect. This is a realist view in which there are objective truths to be discovered in the world. But moral, political, and legal decisions are different from such prosaic examples. So, some have argued that normative statements are mere matters of opinion, lack truth value, and may be little more than non-cognitive expressions of emotion like “SCOTUS decision — Boo!” or “SCOTUS decision — Huzzah!” In other words, there is no objective truth whatsoever to Rep. Ocasio-Cortez’s evaluative claim that SCOTUS is involved in “a dangerous creep toward authoritarianism.”

Constructivists attempt to locate themselves in between realist views and non-cognitivist views. In constructivist views, statements about morality, politics, and the law have independence from individual human beliefs and emotions, and thus they have a claim to being objective. But the truths in these normative domains are not discovered facts, as a realist would claim. The truths are created by a combination of the rules of the game which we construct, and the facts as they occur as we play the game.

What does this have to do with SCOTUS? Rawls makes a distinction between different types of games we can be playing in political arenas. The difference between these games is the nature of the procedures used to identify evaluative truths. One of these games involves imperfect procedures; another game uses pure procedures.

Imperfect procedures start with the assumption that there exists an independent criterion concerning what constitutes a correct answer, but that the practices available to an investigator do not guarantee that the correct answer will be identified. Rawls uses a jury trial as an example. Lawyers use the adversarial method to convince a jury of peers that the accused either is or is not legally guilty. But those who are watching often want a verdict to express not legal guilt but something more — namely did the accused actually act in a way that warrants a conviction or acquittal. As Rawls puts it, in a jury trial using the adversarial method, “[e]ven though the law is carefully followed, and the proceedings fairly and properly conducted, it may reach the wrong outcome.” The existence of the appeals procedure, and the work of organizations like The Innocence Project, suggests that the adversarial method does not guarantee the correct answers.

Alternatively, the major difference with a pure procedure is that a pure procedure does not assume the existence of an independent reality and criterion for what constitutes a correct answer. Instead, the procedure itself, when run, constructs the correct answer. An example of such procedures is a bracket procedure used in professional tennis tournaments. There is no answer to the question “who is the best women’s player at Wimbledon in 2023?” until the tournament is finished. Iga Swiatek may have been ranked number one in the world and the number-one seed at Wimbledon, but before the tournament finished, there was no answer to the question. Now that the tournament is over, we can answer the question: Markéta Vondroušová is the best women’s player at Wimbledon in 2023, despite being unseeded at the beginning of the tournament.

If we want to know whether SCOTUS or the other federal courts are functioning properly or in need of reform, we need to determine what type of procedure governs the practices of the courts and see whether SCOTUS is behaving in accordance with these conceptions. Begin with the following question: Is SCOTUS supposed to be operating in a manner consistent with imperfect procedural justice?

At first glance, that seems to be the case. Decisions made by SCOTUS are supposedly constrained by several independent criteria. First, the content of the U.S. Constitution, especially the Bill of Rights, is supposed to be an independent constraint on their rulings. SCOTUS can’t ignore issues of due process. SCOTUS and the legislative bodies have the ability to shape the nature of due process, but the Fifth Amendment requires that certain rules are followed. Next is the concept of standing which requires a plaintiff to have an interest that is violated, an “injury-in-fact.” This keeps the court from making decisions based on hypothetical cases of legal injury. There is also the concept of stare decisis, the idea that courts should honor precedent to create a predictable and consistent legal system (see here).

A Supreme Court justice that had this many constraints and attended to them should, as Chief Justice John Roberts stated in his confirmation hearing, see their “job as to call balls and strikes.” In other words, the role of a justice is to look at the facts of a case and consider the independent constraints of precedent, statutory laws, and constitutional laws, and merely make a judgment about the truth in the case.

Has SCOTUS lived within these constraints? Some have argued that the Plaintiff Lorie Smith did not have standing in 303 Creative LLC because a) no one asked her to make a website for a wedding between two people of the same sex, and b) no state agency invoked the Colorado Anti-Discrimination Act in legal action against her. She suffered no injury-in-fact. Further, given that Dobbs overturned an almost 50-year-old precedent in Roe v. Wade and SCOTUS overturned decades of precedent in Students for Fair Admissions, Inc. v. President and Fellows of Harvard College, there is reason to believe that the court is not abiding by stare decisis. Admittedly, stare decisis is not an absolute constraint, otherwise the ruling of Brown v. Board of Education would also be problematic. However, given the current willingness to claim that a previous case was wrongly decided despite being upheld by other decisions suggests it is currently no constraint at all.

Next, consider that Marbury v. Madison reinforced the concept of judicial review – the Court can invalidate laws and statutes considered inconsistent with the Constitution. Judicial review is effectively the power to interpret the prevailing legal institutions, especially the Constitution, to justify decisions. So, it is not simply the Constitution, but a justice’s interpretation of that document, that is operative in justifying SCOTUS decisions. Ultimately, a justice issues guidance as to the correct answer according to how their particular interpretive lens (originalism, contextualism, legal pragmatism, etc.) makes sense of the facts of the case. But that means the Constitution is not, by itself, an independent constraint on judicial decisions either.

In short, it is not clear that there are constraints on SCOTUS, so it looks like we should not see it as using imperfect procedures.

Does SCOTUS instead operate according to a pure procedure? It might seem so since we don’t really know the correct answer to a legal question until SCOTUS rules. The process of appealing a lower-court decision, SCOTUS hearing the case, and making a decision settles the matter. Pure. Procedure. So, on this view, SCOTUS in Dobbs, 303 Creative, etc., is functioning as it should, and the basis for criticizing the Roberts Court as radical, activist, illegitimate is questionable, as is the call for reform.

But it is important to note that SCOTUS is not like Wimbledon. Tennis tournaments have rules that are drawn up by one group of people, while another group of people act as judges during matches. The procedure — the bracket — then develops based on the actions of the players as overseen by the judges. This is decidedly different from how the Supreme Court currently operates. Between choosing judicial interpretive lens, interpreting the rules of standing, determining whether stare decisis should be honored, SCOTUS is fundamentally not like these sports tournaments. When a case comes before SCOTUS, the justices can act as both the umpires and the rules committee. They make choices about the relevant rules and their application after the legal competition has begun. In sports, all of these decisions are made and publicized prior to the first match. So, in very important ways, SCOTUS operates in a way that is not even consistent with a paradigm example of a pure procedure. In other words, these are not decisions created by a combination of the rules of the game which we construct, and the facts as they occur as we play the game. They are also constructed by those who are changing the rules as the game proceeds.

If we accept the above, then recent decisions show the Court is functioning as designed, but perhaps not functioning as we want. The decisions are not discovered by an imperfect procedure; they are not created by a pure procedure. Instead, they are the result of something else, such as expression of political power. So, what can be done? When evaluating institutions, Rawls would have us ask “is this how a group of rational individuals, unaware of their own biases or even personal characteristics, would design the rules of a judicial system?” If the answer is “No! This is not the design we want,” will increasing the number of judges fix the problem?  Unfortunately, increasing the number of justices, as Ocasio-Cortez suggests, will not make the court function properly — it will simply shift the balance of power. Instead, what is needed are enforceable constraints on the decisions of justices so that SCOTUS operates more like a sporting event and less like a pure exercise of political power.

 

The author would like to acknowledge the help of his colleague Dr. Dominic DeBrincat, Professor of History.

Posted on

Book Bans, the First Amendment, and Political Liberalism

photograph of banned book display in public library

Book bans in public schools are not new in America. But since 2021, they have reached levels not seen in decades, the result of efforts by conservative parents, advocacy groups, and lawmakers who view the availability of certain books in libraries or their inclusion in curricula as threats to their values. In one study that looked at just the nine-month period between July 1, 2021 and March 31, 2022, the free expression advocacy organization PEN America found nearly 1,600 instances of individual books being banned in eighty-six school districts with a combined enrollment of over two million students. Of the six most-banned titles, three (Gender Queer: A Memoir, All Boys Aren’t Blue, and Lawn Boy) are coming-of-age stories about LGBTQ+ youth; two (Out of Darkness and The Bluest Eye) deal principally with race relations in America; and one (Beyond Magenta: Transgender Teens Speak Out) features interviews with transgender or gender-neutral young adults. 41% of the bans were tied to “directives from state officials or elected lawmakers to investigate or remove books.”

The bans raise profound ethical and legal questions that expose unresolved issues in First Amendment jurisprudence and within political liberalism concerning the free speech rights of children, as well as the role of the state in inculcating values through public education.

What follows is an attempt to summarize, though not to settle, some of those issues.

First, the legal side. The Supreme Court has long held that First Amendment protections extend to public school students. In Tinker v. Des Moines Independent Community School District, a seminal Vietnam War-era case about student expression, the Court famously affirmed that students in public schools do not “shed their constitutional rights to freedom of speech or expression at the schoolhouse gate.” Yet student expression in schools is limited in ways that would be unacceptable in other contexts; per Tinker, free speech rights are to be applied “in light of the special characteristics of the school environment.”

Accordingly, Tinker held that student speech on school premises can be prohibited if it “materially and substantially disrupts the work and discipline of the school.”

The Court has subsequently chipped away at this standard, holding that student speech that is not substantially and materially disruptive — including off-campus speech at school-sponsored events — can still be prohibited if it is “offensively lewd and indecent” (Bethel School District No. 403 v. Fraser), or can be “reasonably viewed as promoting illegal drug use” (Morse v. Frederick). In the context of “school-sponsored expressive activities,” such as student newspapers, the permissible scope for interference with student speech is even broader: in Hazelwood School District v. Kuhlmeier, the Court held that censorship and other forms of “editorial control” do not offend the First Amendment so long as they are “reasonably related to legitimate pedagogical concerns.”

Those cases all concerned student expression. A distinct issue is the extent to which students have a First Amendment right to access the expression of others, either through school curricula or by means of the school library. Book banning opponents generally point to a 1982 Supreme Court case, Board of Education, Island Trees Union Free School District No. 26 v. Pico, to support their argument that the First Amendment protects students’ rights to receive information and ideas and, as a consequence, public school officials cannot remove books from libraries because “they dislike the ideas contained in those books and seek by their removal to prescribe what shall be orthodox in politics, nationalism, religion, and other matters of opinion.”

There are, however, three problems with Pico from an anti-book banning perspective. First, those frequently cited, broad liberal principles belong to Justice Brennan’s opinion announcing the Court’s judgment. Only two other justices joined that opinion, with Justice Blackmun writing in partial concurrence and Justice White concurring only in the judgment. Thus, no majority opinion emerged from this case, meaning that Brennan’s principles are not binding rules of law. Second, even Brennan’s opinion conceded that school officials could remove books from public school libraries over concerns about their “pervasive vulgarity” or “educational suitability” without offending the First Amendment. This concession may prove particularly significant in relation to books depicting relationships between LGBTQ+ young adults, which tend to include graphic depictions of sex. Finally, Brennan’s opinion drew a sharp distinction between the scope of school officials’ discretion when it comes to curricular materials as opposed to school library books: with respect to the former, he suggested, officials may well have “absolute” discretion. Thus, removals of books from school curricula may be subject to a different, far less demanding constitutional standard than bans from school libraries. In short, Pico is a less-than-ideal legal precedent for those seeking to challenge book bans on constitutional grounds.

The question of what the law is is, of course, distinct from what the law should be. What principles should govern public school officials’ decisions regarding instructional or curricular materials and school library books?

A little reflection suggests that the Supreme Court’s struggle to articulate clear and consistent standards in the past few decades may be due to the fact that this is a genuinely hard question.

Political liberalism — the political philosophy that identifies the protection of individual liberty as the state’s raison d’être — has traditionally counted freedom of expression among the most important individual freedoms. Philosophers have customarily offered three justifications for this exalted status. The first two are broadly instrumental: according to one view, freedom of expression promotes the discovery of truth; according to another, it is a necessary condition for democratic self-governance. An important non-instrumental justification is that public expression is an exercise of autonomy, hence intrinsically good for the speaker.

The instrumental justifications seem to imply, or call for, a corresponding right to access information and ideas. After all, a person’s speech can only promote others’ discovery of truth or help others govern themselves if that speech is available to them. Simply having the unimpeded ability to speak would not contribute to those further goods if others were unable to take up that speech.

Yet even if the right of free speech implies a right to access information and ideas, it may be plausibly argued that the case for either right is less robust with respect to children. On the one hand, children generally have less to offer in terms of scientific, artistic, moral, or political speech that could promote the discovery of truth or facilitate democratic self-governance, and since they are not fully autonomous, their speech-acts are less valuable for them as exercises of their autonomy. On the other hand, since children generally are intellectually and emotionally less developed than adults, and also are not allowed to engage in the political process, they have less to gain from having broad access to information and ideas.

Obviously, even if sound, the foregoing argument only establishes lesser rights of free speech or informational access for children, not no such rights. And the case for lesser rights seems far weaker for teenagers than for younger children. Finally, the argument may be undermined by the state and society’s special interest in educating the young, which may in turn provide special justification for more robust free speech and informational access rights for children. I will return to this point shortly.

All the states of the United States, along with the federal government, recognize an obligation to educate American children. To fulfill that obligation, states maintain public schools, funded by taxation and operated by state and local government agencies, with substantial assistance from the federal government and subject to local, state, and federal regulation. As we’ve seen, the Supreme Court has mostly used the educational mission of the public school as a justification for allowing restrictions on students’ free speech and informational access rights inasmuch as their exercise would interfere with that mission.

Thus, the Court deems student speech that would disturb the discipline of the school, or books that would be “educationally unsuitable,” as fair game for censorship.

This is not radically different from the Court’s approach to speech in other public institutional contexts; for example, public employees’ speech is much more restricted than speech in traditional public forums. The combination of the sort of considerations adduced in the last paragraph, together with idea that speech and informational access can be legitimately restricted in public institutions, may lead one to conclude that student expression and informational access in public schools can be tightly circumscribed as long as it is for a “legitimate pedagogical purpose.”

This conclusion would, I think, be overhasty. The overriding pedagogical purpose of the public school does not cleanly cut in favor of censorship; in many ways, just the opposite. Educating students for citizenship in a liberal democracy must surely involve carefully exposing them to novel and challenging ideas. Moreover, mere exposure is not sufficient: the school must also encourage students to engage with such ideas in a curious, searching, skeptical, yet open-minded way. Students must be taught how to thrive in a society replete with contradictory and fiercely competing perspectives, philosophies, and opinions. Shielding students from disturbing ideas is a positive hindrance to that goal. This is not to deny that some content restrictions are necessary; it is merely to claim that the pedagogical mission of the public school may provide reason for more robust student free speech and informational access rights.

But what about conservatives’ objections — I assume at least some of them are made in good faith — to the “vulgarity” of certain books, irrespective of their intellectual content? Their determination to insulate students from graphic descriptions of sex might seem quixotic in our porn-saturated age, and one might think it is no worse than that. In fact, insofar as these objections derive from the notion that it is the job of public schools to “transmit community values,” as Brennan put it in Pico, they raise an important and unresolved problem for political liberalism.

Many versions of political liberalism hold that the state should strive to be neutral between the competing moral perspectives that inevitably exist in an open society.

The basic idea is that for the sake of both political legitimacy and stability, the state ought to be committed to a minimal moral framework — for example, a bill of rights — that can be reasonably accepted from different moral perspectives, while declining to throw its weight behind one particular “comprehensive doctrine,” to use John Rawls’s phrase.

For example, it would be intuitively unacceptable if state legislators deliberated about the harms and benefits of a particular policy proposal in terms of whether it would please or enrage God, or of its tendency to help the public achieve ataraxia, the Epicurean goal of serene calmness. One explanation for this intuition is that such deliberation would violate neutrality in employing ideas drawn from particular comprehensive doctrines, whether secular or religious, that are not part of that minimal moral framework with which most of the public can reasonably agree.

If state neutrality is a defensible principle, it should also apply to public education: the state should not be a transmitter of community values, at least insofar as those values are parochial and “thick,” rather than universal and “thin.” Concerns about children’s exposure to graphic depictions of sex may be grounded in worries about kinds of harm that everyone can recognize, such as psychological distress or, for certain depictions, the idea that they encourage violent sexual fantasies that might later be enacted in the real world. But conservatives’ worries might also be based in moral ideas that don’t have much purchase in the liberal moral imagination — ideas about preserving sexual purity or innocence, or about discouraging “unnatural” sexual conduct like homosexuality. These ideas, which are evidently not shared by a wide swath of the public, do not have a place in public education policy given the imperative of state neutrality.

Unfortunately, while perhaps intuitively compelling, the distinction between an acceptably “minimal” moral framework and a “comprehensive doctrine” has proved elusive. For example, are views about when strong moral subject-hood begins and ends necessarily part of a comprehensive doctrine, or can they be inscribed in the state’s minimal moral framework? Even if state neutrality can be adequately defined, many also question whether it is desirable or practically possible. Thus, it remains an open question whether the transmission of parochial values is a legitimate aim of public education.

Public educators’ role in mediating between students and the universe of ideas is and will likely remain the subject of ongoing philosophical and legal debate. However, this much seems clear: conservative book bans are just one front in a multi-front struggle to reverse the sixty-year trend of increasing social liberalization, particularly in the areas of sex, gender, and race.

Posted on

Underexplained Concepts in the Abortion Debate

photograph of pro-life protesters with religious abortion signs

In my previous column, I tried to demonstrate that some concepts in the abortion debate, namely, the concept of moral personhood and potential lives, are often underexplained. When we analyze these concepts, we find that they are ambiguous and our attempts to define them may significantly shape our views about abortion.

The concepts I focused on previously were strictly about debates regarding whether abortion is immoral. Today, I want to change my focus and strictly consider concepts that apply directly to debates about whether or not abortion ought to be legal.

The Purpose of the Law

A common refrain among advocates of reproductive choice is that you cannot eliminate abortions by outlawing them, you merely eliminate safe abortions. Indeed, some data even suggest that banning abortions does not reduce their occurrence.

So advocates of reproductive choice instead claim that, to minimize abortion, we should favor policies that enable things like easy access to contraception and comprehensive sex education.

In other words, we reduce the number of abortions by reducing the number of unwanted pregnancies.

What’s so odd about this argument is that it should be effective. If one’s goal is to eliminate as many abortions as possible, then surely one should implement the policies that reduce it and minimize the harm they produce when one inevitably occurs. Yet this argument seems to have little, if any, purchase with anti-abortion advocates. Why might this be?

To demonstrate what’s going on here, consider a thought experiment. Imagine that we lived in a society where murder was never formally outlawed. Despite this, the murder rate has never been higher than that found in other nations – the fear of social sanction and our general apprehension towards harming others have kept most people in check. Yet our lack of anti-murder statutes has been something of a national embarrassment. You decide to join a campaign to make murder illegal.

Suppose you’re canvassing as part of the campaign. You knock on my door. I open it and listen to your spiel. However, part way through I begin to shake my head and say the following:

“No, no, no. You’ve got it all wrong. We should simply invest in conflict mediation, anger management, and self-defense courses. This is how we prevent murder.”

There’s something head-scratching about my response. You do not want murder to be illegal merely to reduce its occurrence. Rather, you seem to be motivated by a concern for what murder being legal says about our society.

How can we explain this discrepancy? Declaring something as illegal accomplishes (at least) two things. First, it coerces people to avoid doing it. In this regard, the success of a law is determined by the law’s consequences. Prohibition of alcohol in the United States was a failure due to its consequences; it led to organized crime and did not prevent drinking.

Second, declaring something as illegal sends a message. Namely, it sends the message that some particular act is unacceptable. Call this the law’s expressivist content. For instance, in Texas v. Johnson, the Supreme Court considered the constitutionality of a law outlawing the burning of a U.S. flag. Surely, the reason for this law was not to curb the negative consequences of burning the flag. Rather, the law was created to send a message: “desecrating” objects like the flag is unacceptable.

Now we can see why arguments appealing the consequences of outlawing abortion ring hollow to anti-abortion advocates.

Their concern isn’t about the consequences of whether abortion is legal. Instead, their concern seems to be primarily about the expressivist content of allowing abortions.

They view abortion as a grave moral ill and, thus, object to the tacit endorsement of abortion as acceptable that is communicated by the procedure remaining legal.

This last point, however, may come too swiftly. Even here this argument may be collapsing a significant difference together. The argument uses a moral claim and infers a legal claim from it. But this does not necessarily follow.

Morality & Legality

Even though we often justify our laws on the basis of morality, morality and legality often come apart. For instance, few would think you did something seriously wrong if you, after checking carefully, ran through a red-light at 2 AM with no other drivers on the road. However, you would be doing something illegal. One might argue that tobacco companies engage in an immoral practice by selling their products – they knowingly sell goods that are physically addictive and cause serious harm to the customers. Yet, unless a tobacco company is violating regulations, they are not acting illegally.

But why allow for this difference? Why shouldn’t our laws wholly overlap with our morality? First, the law may appeal to factors that morality does not consider. For instance, laws often consider the costs and effects of enforcement. Most think that lying for selfish purposes is immoral. But enforcing a law against this would be disastrous – it would require huge disruptions of our personal lives, it would result in citizens having to report suspected liars to the police, and a very large amount of money spent on investigating these allegations.

Further, the law is (or ought to be) concerned about citizens’ ability to endorse it and justify it to others. The philosopher John Rawls introduces two concepts which are helpful to consider here: what he calls overlapping consensus and public reason.

Overlapping consensus is reached when citizens in a country are able to endorse the same laws for their own personal reasons, even if those personal reasons differ.

Consider again murder being illegal. This is a law that anyone can endorse – Abrahamic religions have commandments against killing, Buddhists think there is a general prohibition on violence, Atheists may think that murder is bad because of the suffering it causes, etc. Despite having different reasons for accepting the law, each endorses it as legitimate. Thus, overlapping consensus has been reached.

While overlapping consensus deals with our private reasons for endorsing the law, the concept of public reason deals with how we persuade others in political debates. According to Rawls, public reason requires, in part, presenting justification for laws and policies using claims that anyone can accept as true. This does not forbid anyone from, say, having a religious reason for wanting a particular law. But rather, the idea is that public discourse should rely on values and reasoning that others view as legitimate values and acceptable ways of reaching conclusions.

This demonstrates two significant problems. First, as my previous column suggested, our views on abortion depend on our views regarding numerous philosophically complicated questions. This is a problem for public discourse itself. Second, even we have worked out our own, individual views on these matters, this does not necessarily imply that we ought to outlaw abortion.

When determining what should and should not be illegal, we need to consider far more than what is moral. We need to consider what the purpose of a particular law actually is, the costs of enforcing it, or what happens if we do not allow the state to intervene.

Further, we must ask ourselves whether the reasons that we publicly present to defend our views are reasons that anyone could accept, or if they rely on some framework that other citizens do not endorse.

In closing, I want to note an under-appreciated line of thought. Judith Jarvis Thomson in “A Defense of Abortion” presents what are perhaps the most influential philosophical arguments about the morality of abortion (helpfully summarized here by Giles Howdle). Near the end of this article, Thomson distinguishes between what she calls Good Samaritans and Minimally Decent Samaritans. Good Samaritans are those who will go out of their way, perhaps at significant cost, to aid others. Minimally Decent Samaritans are those who do something to help others, even if it is not much.

Morality often requires us to be Minimally Decent Samaritans. Morality may sometimes call for us to be Good Samaritans. The law, on the other hand, does not even require us to be Minimally Decent. Indeed, unless we’re characters in the finale of Seinfeld, the law does not compel us to aid others in need even if it would be monstrous for us to remain bystanders. Making abortion illegal would be an extreme exception – it would require the pregnant person to make many significant sacrifices for the sake of saving another’s life.

Posted on

Acquitted but Not Forgotten: On the Ethics of Acquitted Conduct Sentencing

black and white photograph of shadow on sidewalk

On March 28th, 2022, the House of Congress was addressed by Congressman Steve Cohen (TN-09), Chairman of the Judiciary Subcommittee on the Constitution, Civil Rights, and Civil Liberties. The topic of the address was the legality of a practice called Acquitted Conduct Sentencing. Acquitted Conduct Sentencing is the practice of a judge increasing the penalty for a crime based on facts about the defendant’s past — specifically, facts about crimes the defendant was charged with, but later acquitted of. Perhaps surprisingly, such a practice is not only legal, but relatively common. For example, in 2019 Erick Osby of Virginia was charged with seven counts of criminal activity related to the possession of illegal narcotics and firearms. He was acquitted of all but two of the charges, which should have resulted in a prison sentence of between 24-30 months. The district court trying him, however, estimated a range of 87-108 months due to the five other charges of which he was acquitted. Osby ended up receiving an 84-month (7-year) prison sentence.

In his remarks, Congressman Cohen, along with the co-author of his bill, Kelly Armstrong (Congressional Representative for North Dakota at large), presented arguments to Congress for making Acquitted Conduct Sentencing an illegal practice. The reasoning is fairly straight-forward: if someone is charged with a crime but later acquitted, that acquittal seems to say that they cannot legally be punished for that particular charge. But when sentences are expanded — even, in some cases, tripled — fines are raised, or obligatory service is extended, due to the charges the defendant has been acquitted of, it certainly seems as though the acquittal was meaningless. Cohen’s argument, then, is clear: if we acquit someone of a charge, they should be fully acquitted, meaning those charges should not have any bearing on the sentence handed down to the defendant.

Still, the question of acquitted conduct sentencing is not quite as straightforward as that. Juries and judges need to make their decisions on a host of factors, some of which have to do with facts known about the defendant’s character as well as predictions about how likely they may be to reoffend. These are not easy decisions to make, and it is further complicated by the ambiguity of what counts as legally-admissible evidence. Acquitting someone of a charge does not entail that no facts relevant to the original charge can be used in the trial. In many cases, it is difficult to say how such evidence should be treated. For example, a charge that was acquitted because of police mishandling evidence may be discussed during witness testimony. That testimony, and facts about the defendant’s character and behavior, seems (at least in some cases) hard to ignore when considering fair and effective sentencing for other charges. Acquittal, after all, does not mean that the defendant is not guilty of the crime, only that they cannot be legally charged for it. This could be for a variety of reasons.

Of course, we know that there are many instances of people being charged with a crime that they are innocent of. Mistaken charges happen all the time. Judges and juries may be privy to the original charge, and the later acquittal, but may not know the reason for the acquittal. Acquitted conduct sentencing allows defendants in this position to suffer the consequences of someone else’s error. Because the people making these legal decisions often have limited, or at least imperfect, access to all of the relevant information, allowing for acquitted conduct sentencing is guaranteed to allow cases like this to (in some cases, massively) increase the sentences for these defendants.

So, how should we think about the ethics of acquitted conduct sentencing? Purely consequentialist reasoning may lead us to conclude that we should look at the statistics: what percentage of acquittals are due to innocence, and what percentage are due to bureaucratic missteps? Perhaps the answer to this question will tell us whether allowing or prohibiting acquitted conduct sentencing would be expected to generally maximize good outcomes. This of course would be based on the presumption that, if someone is genuinely guilty of the crime for which they are acquitted, then adjusting their sentence in light of any relevant facts of the acquitted charge would be best for preventing future harm. But this assumption may, of course, itself be mistaken.

Instead, maybe a just outcome depends on more factors than simply maximizing happiness or minimizing harm. The idea of fairness as a desirable outcome of justice, for instance, is a popular one. We might think about the issue of acquitted conduct sentencing as a question of where the locus of justice lies: is a procedure of justice fair in virtue of the procedure itself, or is it fair just in case the outcomes of the procedure are generally fair? John Rawls, one of the most influential political philosophers of the modern era, argued that what he called perfect procedural justice has two characteristics: (1) criteria for what constitutes a just outcome of the particular procedure, and (2) the particular procedure guarantees that a perfect outcome will be achieved. Of course, such perfection is often unattainable in real life, and we might think that the best we can aim for is imperfect procedural justice: where criteria (1) is met, but the procedure cannot guarantee a perfect outcome. Can our current sentencing procedure meet Rawls’ first characteristic? Does it give us an idea of what counts as a just outcome of sentencing? The answer is unclear.

Further, we might question whether outcomes are relevant at all for justice. As a pluralistic society, we might expect there to be wildly-differing views about what counts as a fair outcome. But what counts as an impartial (if not fair) procedure is likely less controversial. For example, when healthcare resources are very scarce, some institutions use random-lottery (or weighted-lottery) decision procedures to determine who gets the resources and who does not. Even if the outcome seems “unfair” (because not everybody who needs the resource will receive it) it is hard to contest that everyone had a fair shot at the prize. Not everyone agrees that lotteries are just procedures, but they at least appear to be impartial. Perhaps this is enough to secure procedural justice? The view that the procedure alone, and not the outcome, determines the fairness of a procedural process, is what Rawls calls pure procedural justice.

Is the procedure of acquitted conduct sentencing fair? Perhaps an easier question: is it impartial? Likely not. After all, implicit (or explicit) bias can easily result in someone being charged for a crime they did not commit. Those who are members of marginalized groups, then, have a much higher risk of having their sentences expanded due to crimes they did not commit. The procedure is far from impartial, and so the likelihood that it could be a part of a just procedural process appears to be low. While we certainly want judges to have as much relevant information on a case as possible when handing down a sentence, perhaps we can agree with Congressman Cohen that acquitted conduct sentencing is not the way to accomplish this goal.

Posted on

The Social Justice of Copyrights and “Public Domain Day”

photograph of Duke Ellington record

In addition to starting a new calendar year, January 1st marks “Public Domain Day” when copyright restrictions expire for a new batch of artworks, thereby allowing new audiences to view them more easily and new artists to adapt them without needing special permission from the copyright holder. This year, the United States saw certain works from Buster Keaton, Gertrude ‘Ma’ Rainey, Duke Ellington, Virginia Woolf, Agatha Christie, and more enter the public domain, including the classic jazz song “Sweet Georgia Brown” and F. Scott Fitzgerald’s famous book The Great Gatsby.

On the one hand, it might seem like increasing accessibility to cultural artifacts is simply obviously good; given how many high school English classrooms rely on battered copies of Fitzgerald’s story, for example, we can see immediate benefits (both aesthetic and practical) to making it easier and cheaper to purchase new books. But, taken to its logical conclusion, this kind of argument seems to suggest that it might always be necessary for artworks and artifacts to be so accessible. If Gatsby really is so valuable, and if it is so embedded within American culture that it is often called “the great American novel,” then why should Americans have had to pay to read it in the first place? Put differently: why is The Great Gatsby only just now entering the public domain?

In brief, the concept of a copyright offers two related basic protections:

  1. It ensures that artists are compensated for the work that they perform, in a way that
  2. Ensures that society will continually benefit from the work of new artists (who, following from (1), will feel free to pursue their art).

This is why, for example, the Constitution specifically grants Congress the power to “promote the progress of science and useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries.” Basically, in theory, copyrights work to level the social playing field a bit so that artists can (at least potentially) enjoy sufficient financial security to be able to practice their art. In effect, this makes copyrights a matter of social justice, since the people who benefit from these protections the most are precisely those from less-affluent or otherwise disadvantaged backgrounds. Although F. Scott Fitzgerald was not exactly socially disadvantaged, the person aiming to write the next great American novel could easily be discouraged from doing so without the hope of protected financial recompense for their labor offered by the copyright system. That is to say: aspiring writers might instead spend their energy towards non-artistic ends if their Gatsby was to simply immediately enter the public domain without helping the writer to, say, buy groceries.

To illustrate, imagine two people who both have an interest and talent for music: Thomas is born to a wealthy family in Hollywood, while Susan grows up in a lower-middle-class family in the Ozarks. Even if copyrights don’t exist, Thomas still has the luxury to pursue his art to his heart’s content: his family’s wealth offers him a level of comfort that shields him from the risk of “wasting time” on a hobby with no guarantee of compensation. The same cannot be said of Susan so easily: while she might still have plenty of personal reasons for playing music on her own, if the realities of her social position, say, require her to work a full-time job in order to provide for basic necessities, then she would be taking on considerable risk to herself if she instead chooses to devote her time to her art without any real guarantee that her music could offer her a profitable career. In principle, copyright laws offer Susan the promise of some financial protection such that if her art ends up becoming profitable, then she will be able to uniquely enjoy the monetary fruits of her labor without other artists being allowed to copy her work (at least for a time); it’s true that Thomas gets this benefit too, but notice that it doesn’t really affect him — he already had the financial protection to do as he liked with his art in the first place.

So, philosophically speaking, copyrights serve as a mechanism to help underwrite the kind of equality that John Rawls talks about with his first principle of justice: in explaining his view of a free and fair, egalitarian society in A Theory of Justice, Rawls argues that “each person is to have an equal right to the most extensive total system of equal basic liberties compatible with a similar system of liberty for all.” Insofar as copyrights can serve to more fairly distribute opportunities to develop artistic skill and create artworks, they might be thought of as components of a just society. Without protections like this in place, it would become, in principle, roughly impossible for anyone not born into privilege to pursue a career in the arts.

It’s worth noting that this is also why artists cannot copyright “generic concepts” or natural elements of normal life: a copyright is only valid for unique artistic creations. In mid-2020, the estate of Sir Arthur Conan Doyle sued Netflix over the depiction of Sherlock Holmes in its film Enola Holmes; while many of Doyle’s stories involving the character of Holmes have entered the public domain, they all tend to present Holmes as a generally cold and unemotional person. Because it is Doyle’s later stories (that are still under copyright) that see Holmes display more warmth and kindness, the caring demeanor the detective shows his younger sister in the Netflix film provoked the copyright-holder to sue. However, the generally-ridiculed lawsuit was settled out of court in December, presumably because “warmth and kindness” are hardly unique artistic creations.

But this also evidences the problem with the other side of copyright laws: artworks are importantly different than commodities or other products for sale. Fitzgerald and Doyle weren’t just “doing their jobs,” for example, when they wrote The Great Gatsby and the Sherlock Holmes stories: they were effectively contributing to the cultural fabric of our society and the artworks that we collectively use to texture our social fabric with shared points of understanding and reference. It might be argued that, just as “warmth and kindness” are ubiquitous to the point of being un-copyrightable, the cultural familiarity of a character like “Sherlock Holmes” is (or is becoming) similarly un-copyrightable.

Such is the argument for “Public Domain Day.” Only the most radical defenders of the public domain would argue that copyrights are, in principle, problematic: indeed, artists both need and deserve to be secure to create their art (consider also: how else might audiences expect to come by new art to appreciate?). However, over time, the sedimentation of individual artifacts into the cultural consciousness makes a unique property claim on them less clearly valid — particularly after the original artist’s death. Though details differ by country, it is common now for copyrights to extend (in general) for either fifty or seventy years after the death of the artist, allowing both the original creator and their dependents to uniquely benefit from the artwork for a limited amount of time before legal ownership of the artifact is distributed collectively.

Rawls also carves out a space for thinking about copyrights in this way within his Difference Principle that allows for some individuals to benefit more than others if that inequality also serves to benefit the least advantaged in society: presumably promoting the further and continued creation of new artworks (as copyrights are designed to do) is just such a public benefit. But once the general welfare is no longer upheld by the existence of a copyright, it would be just for the copyright to dissolve — as indeed we see demonstrated and celebrated each year on Public Domain Day.

(A crucial note: you may have noticed my repeated hedging in previous paragraphs as I have defended copyright law “in principle” or “philosophically.” This is because the actual practice of copyright law in the United States is fraught with problematic and unfair issues that Rawlsian principles of justice would struggle to support. Indeed, the extension of copyright terms seen in the last few decades, the corporate interests apparently motivating such legislation, and other threats to a shrinking public domain (as well as unique questions posed by new forms of art and media) are all issues that deserve both philosophical and legislative attention in a way that is far more complicated than the simple picture I’ve sketched in this short article!)

Still, copyrights play an important part for anyone looking to protect the financial interests they have bound up in their art; for the rest of us, Public Domain Day grants us the green light to continue bearing back into the past to bring it forward into today.

Posted on

Excessive or Necessary? Prosecutorial Discretion in Pursuing Legal Charges

photograph of courtroom

District Attorneys (DAs) in the United States get to decide which cases their offices will pursue. For the most part, there is nothing beyond pressure from voters and other public offices to provide any external impetus to a DA’s decision regarding whether to take an accused person to trial. The results of prosecutorial discretion throughout US history are decidedly mixed. Throughout the Civil Rights movement (1954-68), prosecutors in the American South routinely refused to prosecute white offenders for racist violence and discrimination against Black people. Within the last decade in the American Southwest there have been threats and attacks against Latino immigrants which have gone unprosecuted, or in which the prosecutor has not requested incarceration for the defendant.

Recently, prosecutors in many jurisdictions have announced their intention not to bring charges against recreational marijuana users, or people violating strict heartbeat-style abortion laws in states like Georgia. These prosecutors deem the relevant laws unjust, either for the disparate impact on non-white citizens or their excessive infringement of sexual autonomy. From all these examples it is clear that prosecutorial discretion can be used both to circumvent just and unjust laws alike. 

But not all legal professionals agree that DAs do, or should, have such wide latitude. In Virginia, Arlington County’s Commonwealth Attorney Theo Stamos has claimed that she has no choice but to prosecute marijuana possession so long as the law remains unchanged. Also in Virginia, that state’s Supreme Court judges have ruled against Norfolk’s Commonwealth Attorney Greg Underwood in his claim that he has complete latitude regarding whether to prosecute crimes in his jurisdiction.

The discussion of prosecutorial discretion touches on the philosophical debate between generalism and particularism. Broadly, this debate is a question about whether (moral) decisions ought to be made on the basis of general principles, or rather particular situations. Professor Jonathan Dancy is one of the most prominent champions of the particularist viewpoint. A given feature of an action—that it’s a lie, for instance—may some times count against doing it but may also sometimes count in favor of doing it. You shouldn’t lie under oath, but you should lie when playing poker. On this basis, particularists argue that there can be no general moral principle to the effect of “do not lie.” The same reasoning is meant to apply to any other potential moral principle. 

Generalists, on the other hand, claim that moral decision-making proceeds by applying general rules to specific situations. Hence, given that there is a moral principle to the effect of “do not lie,” you should neither lie under oath nor when playing poker. The moral philosophy of Immanuel Kant is paradigmatically generalist, as seen in his Categorical Imperative: “Act as if the maxim of your action were to become through your will a universal law.”

Paul Woodruff brings the generalist-particularism debate to the topic of justice and the law in his book The Ajax Dilemma. In the chapter on justice, he notes that thinkers as august as John Stuart Mill and Aristotle have abandoned an abstract, principled notion of justice for a pragmatic, particularist one. He contrasts this account to those of a more abstract and generalist type due to thinkers like Plato and John Rawls. Especially concerning Rawls, Woodruff cautions against equating justice with fairness. Justice is everyone getting their due treatment, whereas fairness is everyone getting the same treatment. While these may sound the same at first blush, Woodruff contends they are in fact worlds apart. Justice and giving people their due requires careful thought and judgment concerning particulars; whereas fairness and giving everyone the same treatment only requires rote execution of rules. 

Attorneys like Theo Stamos have a more generalist approach, one that is plausibly interpreted as treating justice like fairness. The law is the law, and every person must receive exactly the same treatment under the law. An approach like this sees the exercise of discretion as unfair because it means different people are treated differently. On the other hand, attorneys like Greg Underwood have a more particularist approach, one that is plausibly interpreted as treating justice as giving people their due. Here the exercise of discretion allows the peculiarities of a given case or jurisdiction to enter the decision-making process. If the enforcement of a law disproportionately affects the non-white community, discretion allows an attorney to effectively neutralize that law. If there are mitigating circumstances in a particular case, discretion allows the attorney to seek a lesser sentence or drop charges entirely. 

Allowing people to exercise judgment and discretion always creates the potential for malfeasance, dereliction, and oppression. However, it is also what creates the potential for mercy, compassion, and resistance. The solution to the possible pitfalls of prosecutorial discretion is not to limit a DA’s ability to exercise judgment, but rather to carefully scrutinize candidates for the office and elect individuals of experience and integrity.