(Posted during the Social Ontology 2018 Conference, hosted at Tufts) We live in a positivist culture in which many smart people hold fairly simple views of science and believe that all rigorous thought is scientific. Their objection to highly abstract conceptual questions and to questions of value (moral, political, or aesthetic) is that these matters cannot be scientific; hence progress is impossible. Endless debate must result from the brute fact that we hold different opinions.

But we must figure out what to value and what to believe about conceptual issues. Given our cognitive and moral limitations as individuals, our best way to think about such matters is with other people–learning from their perspectives and testing our beliefs with them. Words are not our only tools for thinking together–mathematical notations, diagrams, images, music, and bodily movements also work–but words are awfully helpful and usually play a role even when we make heavy use of the alternatives.

Therefore, human beings talk about conceptual and normative matters. We always do, everywhere and in every era. But a literal conversation has drawbacks. Since an actual, oral dialogue must involve a small number of people, the cognitive resources are limited. It lasts for a finite amount of time–too brief to address all the relevant questions and issues. And it proceeds in a linear fashion, with one comment or question occupying attention at any given moment, followed by the next one. Although people may make discursive moves like saying, “Let’s go back to your earlier point P, because I disagree with it,” the participants can barely explore the whole network of potentially connected ideas. A conversation is one walk through one part of the network.

A discipline like philosophy is an effort to improve the conversation by institutionalizing it. Many people can participate in a discussion that is organized in the form of journals, books, symposia, and reviews. Participants publish their claims and reasons, leaving them on the record to be picked up by others. They take time to make each point carefully, offering reasons and considering objections. If someone claims P, other people are supposed to read and cite that claim before they say P or not-P. If you criticize P, then other people who begin by believing P are supposed to read and consider your objection to P before they use it as a premise in their arguments. The debate still continues permanently, but it is supposed to become increasingly organized and refined in a process that is just as cumulative as a “normal science” is. Moreover, the strictly philosophical debate is not insulated from other intellectual work but is constantly informed by developments in the sciences, humanistic thought, and actual events in the world.

This is all idealized. I am perfectly aware that not everyone can participate in a professional discipline’s discussion; in fact, the vast majority of human beings are excluded, for a whole range of reasons. Nor would everyone want to join even if that were easy for them. Those who participate act imperfectly, showing too much deference to certain authorities, demonstrating group-think, etc. And ethics (in particular) still suffers from myopia about cultural diversity and empirical data that Owen Flanagan well describes in his new book, The Geography of Morals: Varieties of Moral Possibility.

Still the ideal has significance as a heuristic. It draws our attention to Robert Merton’s four CUDOS norms, which he developed for science (per Wikipedia):

• Communalism all scientists should have equal access to scientific goods (intellectual property) and there should be a sense of common ownership in order to promote collective collaboration, secrecy is the opposite of this norm.

• Universalism all scientists can contribute to science regardless of race, nationality, culture, or gender.

• Disinterestedness according to which scientists are supposed to act for the benefit of a common scientific enterprise, rather than for personal gain.

• Organized Skepticism Skepticism means that scientific claims must be exposed to critical scrutiny before being accepted.

These norms also apply to philosophy, and we can add more values, such as 1) the norm of citing and addressing previous contributions to the same discussion; 2) the principle that academic discussions should ultimately (but not always directly) benefit public life; 3) the value of being permeable and connected to other discussions in other fields; and 4) an affirmative effort to incorporate people and perspectives that have hitherto been marginalized.

See also: is all truth scientific truth?, does naturalism make room for the humanities?, why social scientists should pay attention to metaphysics, and adding democracy to Robert Merton’s CUDOS norms for science.

I’d posit the following definitions:

• Empathy: Feeling a similar emotion in response to someone else’s emotional state. Your friend is mad at her boss because he treated her unfairly. That makes you mad at her boss. Your anger is probably different in texture and intensity from hers, but it’s the same in kind, an imperfect reproduction of her mental state.
• Sympathy: Feeling a supportive emotion in response to someone else’s emotional state that is not the same as that person’s original emotion. She is mad at her boss, so you become sorry for her, or committed to fairness, or sad about the state of the world, or nostalgic for better times–but not angry at her boss. Then you are sympathetic. (NB You can be both sympathetic and empathetic if you feel several emotions.)
• Compassion: A species of the genus sympathy. Another person’s negative emotion causes you to have a specific supportive feeling that is not the same as her emotion: you sincerely wish that her distress would end without blaming her for it.
• Justice: A situation or decision characterized by fairness, goodness, rightness, etc. (These are contestable ideas and may be in tension with each other.) The English word “just”–like dikaios in classical Greek–can be applied either to a situation or to a person who cares and aims for justice.

There is an old and rich debate about which character traits and subjective states are best suited to pursuing justice. One answer is that you should be a just person, one who tries to decide what is fair or best for all (all things considered), who desires that outcome, and who works to pursue it. A different response is that we are not well suited to defining and pursuing justice itself. It’s better to cultivate other emotions, such as empathy, sympathy, compassion–or loyalty, aversion to harm, or commitment to specific rules–in order to deliver more just outcomes, all things considered.

I haven’t yet read Paul Bloom’s new book, Against Empathy, but based on this interview and other secondary material, I take it to be an empirically-based contribution to this debate. Bloom marshals evidence that empathy is a highly unreliable guide to justice, probably more likely to mislead than to inform. We should cultivate justice itself, not settle for a substitute, and certainly not the poor substitute of empathy.

For instance, Donald Trump can make people feel empathy for a small number of individuals whose families were allegedly victimized by undocumented aliens, and then use that emotion to build support for deporting millions of people who have harmed no one. A famous example is Edmund Burke’s outrage at the mistreatment of Marie Antoinette, which obscured any concern for the countless people tortured, executed, or “disappeared” by the ancien regime that she represented. (By the way, I happen to respect Burke a lot–and I don’t think it was fair or smart to execute the Queen–but this passage is still a good example of misplaced empathy.)

Empathy can also substitute for justice, as I argued on a visit to Israel in 2013. You congratulate yourself for feeling some version of a suffering person’s emotion and excuse yourself from fixing the problem.

Compassion may be better than empathy. Instead of feeling the same emotion as the other person, you feel a combination of beneficence and equanimity that may be a more reliable guide to acting well. But it’s possible that compassion only clears the deck for reasoning about what you should actually do.

For its part, justice can be emotional. You can feel a powerful urge to make the world more just. That is helpful insofar as the feeling motivates you and insofar as people obtain genuine insights from our emotions; but it is dangerous because the emotion of desiring justice can be misplaced. You can feel great about improving the world when you are actually harming it.

Justice is also necessarily discursive. You must put into words–at least inside your own head–what is good or fair, and why, and make yourself accountable for that position. Therefore, much hinges on whether we human beings can reason explicitly about justice in ways that improve upon our strictly affective reactions to particular situations.

See also: empathy: good or bad?; the limits of putting yourself in their shoes and looking with their eyes; an alternative to Moral Foundations Theory; structured moral pluralism (a proposal); and how to save the Enlightenment Ideal.