fbpx

Which Justice for Which Person?

The Supreme Court’s 2022 term will soon conclude with two cases testing the limits of speech rights involving online services. One features the internet behemoth, Google. The other spotlights a woman named Lorie Smith, whose small business, 303 Creative LLC, creates webpages for couples celebrating their nuptials.

Google’s case has been decided to its satisfaction. It sought shelter from liability under section 230 of the Communications Decency Act (CDA) against a claim that it failed to censor terrorist communications, which allegedly contributed to the death of a woman. But the Supreme Court punted the section 230 defense, ruling instead that the plaintiffs hadn’t shown that Google had aided and abetted in terrorist activity. Lorie Smith’s case has yet to be decided. Lorie Smith is seeking protection under the First Amendment to shield her from the Colorado Anti-Discrimination Act (CADA) in the face of her refusal to create webpages celebrating same-sex marriages.

What is striking about these cases, from a First Amendment perspective, is that legislation exists to shield the behemoth; no legislation exists to protect Lorie Smith. Lorie Smith’s services have been characterized as so indispensable that she must be compelled to offer her artistic services. Google, which is truly indispensable, can purport to be empowered under section 230 to pick and choose who it will serve, while also not being responsible for the content of the messages it hosts.

Should these differences matter to anyone? If grossly disparate treatment favoring the powerful is a cause for concern, the answer is apparent. Yet the statutory provisions involved in each case allow for no equitable comparisons to be drawn. Even those religious freedom cases, like Burwell v. Hobby Lobby, that take notice of the unique characteristics of closely-held corporations, as opposed to publicly-traded companies, like Google, don’t get to the heart of the matter. The Supreme Court’s allowance for more favorable First Amendment treatment for closely-held businesses still doesn’t articulate the underlying reasons for the injustice in the disparate legislative treatment of Google and Lorie Smith.   

One must turn to basic concepts of justice to explain the intuition that something is amiss. According to the ancients, justice, as a general concept, involves giving each their due. Justice, that most social of virtues, requires achieving the right proportion in a relationship. But as there are different kinds of relationships in human affairs, there are also different models of justice. 

In Aristotle’s Nicomachean Ethics, from which St. Thomas borrowed liberally, two types of justice are identified as being of great importance to lawmakers; these are distributive and commutative justice. On the one hand, distributive justice considers how the goods or burdens within a political community ought to be shared among its members, much like a pie or housework might be divided among members of a family. This type of justice applies to a community, taken as a whole, of which members are its parts.

On the other hand, commutative justice is best suited to assessing interpersonal relationships. If one party wrongly injures another, then the person injured has a right to claim recompense for the harm. Commutative justice prevails in civil cases, such as contracts or torts. By contrast, distributive justice is the proper framework to adopt when apportioning something that just about everyone wants, like more income, or conversely, when apportioning something no one desires, like taxes.   

Aristotle’s basic framework helps to explain why the disparate treatment of these two online providers is particularly unjust. Lorie Smith, a business owner with few clients, has been characterized by the Tenth Circuit Court as the holder of a communal good that is so important to society that she must be compelled to provide her services or else harm will befall those who are denied her talents. This is a distributive justice model. And it plainly doesn’t fit.  

Her business, 303 Creative LLC, does no harm to anyone by not contracting with them, apart from perhaps hurting some feelings. No matter her level of talent, no matter whether her work is artistic or prosaic, she is a readily replaceable service provider from a societal perspective. A person’s refusal to contract might, in certain extreme cases, be a matter of societal concern. But generally, society’s interest in any given contract is properly guided by the commutative justice model. This requires the complaining party to show a concrete injury, and even then, that injury must generally be capable of being monetized to put the injured party back in the position they would have been. It is hard to see how anyone is injured by not being able to contract with 303 Creative LLC. Far worse, it is arguably unjust to treat Lorie Smith as something she is not—a social commodity undeserving of choosing the terms upon which she will offer her services.

Google, on the other hand, is an entirely different kind of provider. Its decisions as to how it will distribute its services matter deeply to anyone who needs access to the internet. Google controls a service for which there are few alternative providers. For the time being, it is virtually indispensable. The distributive justice model presumptively fits Google. Still, the fact it dominates the interactive computer services market doesn’t mean that the distributive justice model is the right one in every case. There are circumstances, as in Gonzalez v. Google, in which Google might be sued for a tort, or another commutative justice claim. And in this type of situation, one could understand why Congress might wish to protect an essential service provider from private liability. Such lawsuits can be a way of indirectly shaping who will get access to a platform, and hence of dictating distributive justice criteria. The distributive good of ensuring virtually universal access to Google might well be judged to outweigh the private claims of individuals, especially where third-party posts are concerned.

But here is the rub with section 230. It is hard to see how Google can, in justice, have it both ways. Google shouldn’t be able to control its distribution in the same way that a small business can decide with whom to contract, especially if it is also immune from liability for materials posted by third parties. Otherwise, Google can claim to be exempt from the demands of both distributive and commutative justice. 

Yet it appears that Congress has done exactly that in section 230 of the CDA. Google is empowered to restrict speech according to its own idiosyncratic standards, while it is also immune from liability for harm that results from its hosting of third-party materials. Perhaps it is no wonder that Justice Thomas, in Biden v. First Amendment Institute, expressed alarm from the unprecedented control over the flow of ideas exercised by a handful of online platforms. Distributive justice as well as free speech concerns of the highest order are involved in any policy regulating companies like Twitter or Google.  

One might wonder what ends are served where section 230 and CADA are incongruous with basic principles of justice.

Now consider anew the irony of the disparate treatment of Google and Lorie Smith. In the topsy-turvy world of patchwork legislation, Lorie Smith is more likely to be treated as an indispensable public accommodation that cannot refuse her services than is Google. More than that, Google, an entity whose business model is premised upon facilitating communication globally, can censor speech. Lorie Smith, whose business model is premised upon serving small numbers of people, can, at least according to the Tenth Circuit, be censored in the most perverse way: by being required to generate content she views to be objectionable.

None of this makes much sense when viewed through the lens of justice or free speech principles. But in law, concrete circumstances matter greatly. America’s long history of racism, and the specter of Jim Crow laws that resulted in the segregation of black populations in places of public accommodation, haunt our legislation and jurisprudence. The concern that naturally arises is this: if Lorie Smith were permitted not to provide her services to celebrate same-sex marriages, would that permit other vendors not to sell their wares to black customers? Would the shameful days of racial segregation return absent the heavy hand of government in compelling small businesses to enter contracts contrary to their principles?

In cases involving distributive justice, racial considerations are generally illicit. And this is precisely why Harvard and the University of North Carolina are unlikely to prevail in the Supreme Court in their quest to continue using race as a consideration in their admission processes. Many universities wish to pick and choose whom they will serve by taking skin color into account. Moreover, these same institutions control goods that are in the public interest, given they are gateways to leadership positions. While Harvard and others might argue that no one has a right to be admitted, that analysis skews the underlying distributive issue. The pertinent social question is whether Harvard or the UNC ought to be distributing their scarce resources with race in mind, even if their intentions are good ones.

Yet intentions still matter in our assessments about justice. Those who intend to do harm to others, and in fact do harm, should be held to account. Where a pattern of individual behavior, if looked at only through the lens of commutative justice, becomes a distributive justice issue, where individuals in need of goods and services are repeatedly and systemically denied, justice could very well favor redressing the imbalance. But there is little evidence to believe that this is the case with Lorie Smith’s refusal to use her services to celebrate same-sex marriages. Moreover, the justice of Lorie Smith’s case should be weighed by considering whether Christians and religious believers, celebrating sacraments, could be refused services by small businesses that do not wish to condone their celebrations. Fairness requires these circumstances to be treated the same.

Of course, laws can be poor approximations of justice, while still being legal. Our laws need not be created with ideas of distributive or commutative justice in mind, nor can a judge typically interpret laws based upon the dictates of justice. Instead, models of distributive and commutative justice provide evaluative criteria for assessing whether certain outcomes are unfair. Most are apt to think, for instance, that it is unfair for a county to foreclose on the home of a 93-year-old woman to claim a debt of approximately $12,700 and then not to give her any of the proceeds that exceed the amount owing. This case, also to be decided in the spring of 2023, is Tyler v. Hennepin County. It, too, is a good example of commutative justice.

Yet one might wonder what ends are served where section 230 and CADA are incongruous with basic principles of justice. Where CADA serves primarily to redress the hurt feelings of certain protected classes, it suppresses opinions. Where Google restricts access to material it deems objectionable, even when that material is constitutionally protected, it censors speech. Section 230 and CADA can work for the same ends. But these ends may have little to do with justice.

Lorie Smith might prevail, just as Google has done. But when it comes to distributive justice, let’s look long and hard at the double standard of insulating Google in refusing its services, while characterizing mom-and-pop shops as indispensable public accommodations, which must enter into contracts that they would rather not.

.