Liberal Tolerance and Seeing Others as Agents
On a Hidden but Still Fundamental Postulate of Liberalism
Tolerance is generally understood as one of the foundational principles of liberalism. Tolerance is tied to the acknowledgment that we have no prima facie reason to mind other persons’ business. We have no reason to oppose their belief about what matters or is true, their conception of the good life, to their practices in daily life. For sure, tolerance cannot be unlimited. Reasons for not interfering are only prima facie. We may want to prevent practices and beliefs that we think have adverse effects on other persons who do not endorse these beliefs or participate in these practices. We may also be reluctant to tolerate persons and ideas that are themselves intolerant. But the general idea remains. Liberal tolerance indicates that except in the case we have decisive reasons to the contrary, we should not interfere with others’ beliefs, practices, and choices.
Sculpture Für Toleranze, by Volkmar Kühn, Gera, Germany (By Hans-Peter - originally posted to Flickr as für toleranz, CC BY 2.0, https://commons.wikimedia.org/w/index.php?curid=6039515)
If you accept liberal tolerance, then you recognize that what you think of other persons’ ways of life, whether or not you approve of them, is irrelevant. Tolerance is not approval; it is refraining from interfering with others, especially when we disapprove of their practices. Tolerance is by many accounts a very demanding ideal. Evolutionary and moral psychology strongly suggest that it is not hardwired. Our ancestors have evolved within small and homogenous communities where behavior was tightly monitored by informal but still constant social pressure. Because of this, humans have evolved a strong disposition to sort out individuals in terms of a us/them logic based on whether one belongs to our group and whether they have the same practice as us, and to punish those who dare not conform to the prevailing practices. Tolerance is in this sense a modern invention that could not have emerged but in societies with a minimal level of diversity, where people do not worship the same god (or not in the same way), do not share the same purposes, or do not follow the same norms. A recurrent theme in the writings of many liberal writers is that humans are not finely tuned to live in the “Great Society.” The “strain of civilization,”[1] a co-product of any diverse society essentially consisting of abstract and impersonal relationships, partly finds its origins here. Tolerance demands that we refrain from acting based on our gut feelings and tribalistic dispositions.
The emergence of the concept of tolerance, the fact that an increasingly large number of persons recognize it as a normative ideal, and the fact that this ideal now permeates our social morality can be interpreted as a genuine example of moral progress. Still, as I said above, there may be relatively clear cases where tolerance must be limited. There are strong reasons to not tolerate practices that cause harm to people who are not voluntarily taking part in them, for instance. We may have good reasons to not tolerate practices that are causing animal suffering. There are however cases that are less straightforward because they don’t involve direct harm (in which case tolerance is not warranted) but are not merely due to the disapproval of others’ ways of life. What I have in mind here are cases where some may not want to tolerate practices (including speeches) on the grounds that they could cause some persons to act in a harmful way. Consider for instance the following examples:
Caricatures. We may not want to tolerate practices that caricature persons, ideas, or religious figures, because beyond the fact that these caricatures may be disrespectful, they can have adverse consequences, e.g., by causing violent behavior from or directed toward those whose ideas are caricatured.
Pornography. We may not want to tolerate the production and consumption of pornography on the ground that it favors representations and behaviors that are harmful to women.
These two examples have in common a causal structure. There is a practice P that is thought to cause some behavior B from some set of individuals I and that has adverse effects E on some other set of individuals I’, including individuals that are not taking part in P.[2] These cases test the limit of liberal tolerance because there is a causal claim that, while the harm is not direct, there is nonetheless a well-identified and normatively relevant relationship between the contested practices and some social consequences that we have decisive reasons to act as to avoid them.
It is worth noting that the problem at stake here is not a problem of externalities, not even of “moral” or “psychological” externalities. One reason for disapproving a practice is that it is contrary to one’s convictions or triggers bad emotions. While the practice does not have a “material” or “tangible” effect on others (as pollution emissions for instance have), its external effects are nonetheless real in the sense that its existence causes a change in some normatively relevant state associated with the affected persons (e.g., happiness, well-being).[3] The problem with the notion of moral/psychological externalities is that it is too shallow to justify interfering with individuals’ behavior. One way to understand liberal tolerance is to view it, precisely, as the claim that the existence of moral/psychological externalities does not provide a sufficient reason to interfere with a practice.
In Caricatures and Pornography, however, the reason justifying the interference is that the practice causes (or may cause) some persons to act in a way that is harmful. If you grant that this is a sufficient reason, then the scope of liberal tolerance is considerably restricted. Virtually every practice or way of life can be suspected, at least under some circumstances, to affect in one way or another the behavior of persons in such a way that it leads to more or less severe harm. That indicates that the ideal of tolerance has no normative weight if it is not supported by a conception of moral agency according to which persons are autonomous beings whose behavior cannot be reduced to causal relationships with their environment. This is well expressed by the philosopher Gerald Gaus:[4]
“The Great Society only becomes possible when individuals are understood to be morally autonomous in the sense that they can put aside their fantasies, perversities or foolish notions and respect the legal personality of others, and are properly held morally accountable if they fail to do so. If my neighbour is morally autonomous, her dangerous and repulsive thoughts, library and VCR collection is not my business, for I can nevertheless expect her to act publicly in accordance with my civil personality. If this is so, then even if some types of people are statistically more likely to commit crimes (say, young males) or are more likely to commit crimes after certain experiences (say, viewing Hustler), we do not seek to eliminate the correlates of crime. For a person is not viewed as simple causal force of his viewing habits or hormones, but as a responsible agent who is not at the mercy of causal forces pushing him this way and that.”
As Gaus notes in the paragraph that follows this quote, to many persons this will seem outrageous. If we know that there is a statistical correlation between the fact that people participate in a practice and harmful behavior, this might seem to provide a sufficient reason to interfere with the practice. Of course, what is relevant here is that the harmful behavior is not part of the practice itself. However, it might be thought that the statistical relation is enough to justify interference. It is clear however that this view entails a complete denial of other persons’ moral agency. They are no longer viewed as autonomous moral agents who are able to subject their behavior to social and moral rules even if that goes against their desires or instinctive proclivities, but as heteronomous creatures whose choices are causally determined by factors outside their control.
Even if this is granted, many may be tempted to bite the bullet. Maybe there are indeed individuals who are not autonomous moral agents. However, presumably, the persons holding such views consider themselves as autonomous moral beings, for otherwise there is no reason to ascribe them any value – in this case, they would just be the causal product of factors that their holders do not control. That implies that they assume a moral asymmetry between them and others: they are moral agents, and others are not. What is at stake therefore in cases like Caricatures and Pornography is more than the normative importance of tolerance as an ideal. It is how we conceive ourselves and others as persons, either autonomous moral agents or heteronomous creatures, and whether we view all of us as equals in this respect.
[1] Karl Popper, The Open Society and Its Enemies, 5e édition (London: Routledge, 2011), p. 167.
[2] Note that behavior B is not part of practice P, which is essential for otherwise there is obviously a reason to interfere with P.
[3] This happens for instance under a counterfactual account of causation. If the practice P didn’t exist, individual I would be happier/better.
[4] Gerald Gaus, “On the Difficult Virtue of Minding One’s Own Business: Towards the Political Rehabilitation of Ebenezer Scrooge,” The Philosopher: A Magazine for Free Spirits 5 (1997): 24–28, p. 26.
"It is clear however that this view entails a complete denial of other persons’ moral agency. They are no longer viewed as autonomous moral agents who are able to subject their behavior to social and moral rules even if that goes against their desires or instinctive proclivities, but as heteronomous creatures whose choices are causally determined by factors outside their control."
This seems to me to be at false dichotomy..
The evidence seems quite convincing that human agents are neither purely autonomous nor purely heteronomous. Instead, agents are indeed able to make choices about behaviour - but that this ability (and often the choices themselves) are also conditioned by factors outside their control.
If we allow that there are good reasons to limit the actions of others - for example in cases of direct harms - then it would follow thereby that we cannot simply rule out (on the basis of some imagined pure autonomy) limitations in cases of indirect harms.