The Power of Ahimsic Communication

“The law of love will work, just as the law of gravitation will work.”  – Gandhi

In parts one and two of this three-part series, I developed a framework for ahimsic (nonviolent) communication (AC) as an alternative to the standard communicative norm of civility. The framework presented for AC offers various categories of resistance to violence, including nonviolent forms of negotiation, compromise, protest, verbal force, verbal distraction, argumentation, and communicative satyagraha (Gandhian nonviolence applied to communication). I also provided a range of real-life examples of successful AC resistance, including the stories of Derek Black, Daryl Davis, James Lawson, and Antoinette Tuff. These examples demonstrate that AC “works” even against neo-Nazis, KKK grand wizards, angry punch-throwing bikers, and active school shooters. In this third and final installment, I will explain why AC works—better than any form of violent communication (VC). My main claim is that AC derives strategic power from its moral power. As various components of my argument are drawn from the nonviolence tradition more generally, what follows will double as a primer on the morality of nonviolence. However, we will also see significant differences in how we must adapt nonviolent strategy and morality to the case of communication.

My overall argument is as follows: Some forms of civility (namely what William Keith and Robert Danisch call weak civility) and some forms of incivility (namely what I shall call righteous incivility) are the best candidates for justified VC. Weak civility (WC) and righteous incivility (RI) use harm as a means to achieve different aims. This shared approach renders them both ineffective. By serving the same positive functions as WC and RI via nonviolent means, AC achieves better ends. This “Hegelian synthesis” thereby vindicates the fundamental hypothesis of Gandhian morality (FHGM): the thesis that the means prefigure the ends, linking moral principle to strategic effectiveness.

The Fundamental Hypothesis of Gandhian Morality

Let’s begin by exploring this hypothesis. To clarify, I regard it as a hypothesis rather than an axiom or a conjecture, since for Gandhi it was more than a starting point or guess. Gandhi’s law of truth is closer to an axiom in his philosophy, whereas the FHGM is a hypothesis in the Gandhian experimental spirit: to be accepted, it must be confirmed, and to be confirmed it must be tested by reason and experiment. Ultimately, Gandhi believed the hypothesis is confirmed. 

The FHGM is a synthesis of two perspectives on morality: (1) the thesis that the ends justify the means (consequentialism) and (2) the antithesis that only the means matter, never the ends (a strict deontology). These two perspectives correspond respectively to two nonviolentist camps: (a) the thesis that nonviolence is justified purely on the basis of strategy (strategic nonviolence) and (b) the antithesis that nonviolence is justified purely on the basis of principle (principled nonviolence). The FHGM maintains that means and ends are inseparable, which corresponds to a third camp (integral nonviolence)—the thesis that nonviolence is justified on the basis of both strategy and principle, which are inseparable.

In defense of principled nonviolence, we might run the standard deontological arguments, making use of stock examples like the organ-harvesting case: most view it as morally wrong to kill one person and harvest their organs, even if doing so were “necessary” to save five others in dire need of transplants. Similarly, one might argue, it is morally wrong to use violence, even when it is “necessary” to reach a just outcome. In response, one might object that in the organ-harvesting case, the potential victim is innocent, and therefore doesn’t deserve to be killed. One might therefore point to a disanalogy: most people restrict violence to cases in which the recipient deserves it, and therefore has waived the right not to be harmed. For reasons to be examined, however, principled nonviolentists avoid thinking in terms of moral desert.  

While Gandhi is often classified as a principled nonviolentist, and while I agree that he is closer to a principled nonviolentist than a strategic nonviolentist, he counts as an integral nonviolentist given my definitions. Let’s examine what Gandhi had to say on the matter.

Inspired by the Bhagavad Gita, Gandhi argued (and Martin Luther King following him) that means and ends are inextricably linked: “The means may be likened to a seed, the end to a tree; and there is just the same inviolable connexion between the means and the end as there is between the seed and the tree” (58). Thus, if I want an apple tree, I must plant an apple seed, and if I want an oak tree, I must plant an acorn. In this way, means and ends are causally connected. This causal connection often comes in the form of what Richard Gregg called moral jiu-jitsu whereby one’s nonviolent stance has a positive effect on opponents and/or onlookers (a concept unfortunately reworked by Gene Sharp into amoral political jiu-jitsu).

Causal connections are contingent. Even so, they establish a tendency that we cannot ignore in decision-making (since we cannot always predict when a tendency will in fact fail to hold). More strongly, Gandhi’s lesser-known watch analogy demonstrates that means and ends are constitutively connected:

If I want to deprive you of your watch, I shall certainly have to fight for it; if I want to buy your watch, I shall have to pay for it; and if I want a gift, I shall have to plead for it; and, according to the means I employ, the watch is stolen property, my own property, or a donation. Thus we see three different results from three different means. Will you still say that means do not matter? (59)

On the usual understanding, consequentialists narrowly construe the end as an isolated point (the watch), arbitrarily cut off from the means by which it was produced (being stolen, purchased, gifted, or borrowed). When we look closely, we see that in full view, the end is much broader, containing the means within it. I’ll therefore distinguish the narrow content (the ends minus the means) from the wide content (the means itself, viewed as part of the ends). The narrow and wide contents together constitute the ends in entirety. Thus, one does not just acquire a watch (the narrow content). One acquires it by stealing, purchasing, receiving, or borrowing (the wide content). Hence, the full end is a stolen watch, purchased watch, gifted watch, or borrowed watch (the combined narrow and wide content). A problematic means, such as violence, therefore infects what would otherwise be desirable ends.

While violence always determines the wide content of the ends, it need not determine the narrow content, and hence need not be noticed or felt by the wrongdoer. In some cases, it is felt, such as when the means includes a moral emotion (e.g., love or hate). In this case, the positive or negative emotional residue appears in the phenomenology of the end, as we will see.  

Due to these causal and constitutive connections, we expect certain kinds of results from certain kinds of means: they are epistemically connected. While we cannot perfectly predict the ends narrowly construed, we can securely predict the wide content that determines their evaluation, namely by selecting the means. As Gandhi observed, this in turn has implications for what we can and cannot control, an understanding of which is essential for our decision to act. We have immediate power over the means, hence over the wide content, but none over the narrow content.

This brings us to a point about success and failure due to Robert L. Holmes: whether or not moral means lead to success and immoral means lead to failure depends on what we mean by “success” and “failure” (163). If the goal is to obtain a watch, we might fail. If the goal is to obtain a watch as one’s own property or gift or loan, then again we might fail. But this precludes theft, and so part of the goal is to avoid obtaining a stolen watch (at least one stolen by us). This we can guarantee by selecting a moral means. And this is all that reflects on our character.

As an analogy, consider a sports match. We praise well-played games—whether wins or losses—over “wins” achieved by cheating. It’s how we play the game that matters, which determines whether it can count as a real “win,” and we can always take satisfaction in playing well even if we “lose.” As an added practical benefit, by focusing on how we play rather than the narrow result, this focus increases the chances of “winning”—a point related to the Stoic dichotomy of control, the Daoist principle of wu-wei, and Mihaly Csikszentmihalyi’s notion of flow, all of which are conducive to personal satisfaction.

There is much more to say about the fascinating and complex relationship between means and ends that cannot be developed further here. Interested readers are encouraged to explore John Dewey on this topic. Here my aim is merely to introduce this mode of moral thought to unfamiliar readers, show how it applies to communication, and demonstrate the implications for AC versus VC.

Turning now to communication, strategic AC maintains that AC is a better strategy than VC, whereas principled AC maintains that AC is morally preferable to VC as a matter of principle regardless of strategy. Integral AC maintains that AC is morally preferable to VC because moral ends require moral means, linking strategy to principle. If this is right, then civility and incivility, insofar as they use harmful means, are poor strategies. By holding to moral means, AC is the more strategic approach.

Let’s put this hypothesis to the test. To do so, we must compare AC to VC. I will limit attention to the only two prima facie candidates for justified VC, to be identified in the next section. I will argue that both candidates ignore AC because they presuppose a false dichotomy. I will then argue that AC has different strategic advantages over each specific form of VC. At first, I will focus on the advantages that arise due to the causal connection between means and ends, which apply even if we (fallaciously) restrict our communicative goal to the narrow content of the ends. But we will eventually return to the stronger constitutive connection between means and ends.

A False Dichotomy

Most of us agree that violence is typically wrong. However, most are unaware of the numerous subtle ways in which common communicative modes are violent. In part one of this series, I argued that VC includes what Keith and Danisch call weak civility (WC), roughly civility in the sense of politeness as a means to silence outcries against injustice or to avoid the discomfort of confronting it. In other cases, violence is overt but condoned as an exception to the general prohibition, namely in service of self-defense, other-defense, or deserved punishment. In communication, this mainly comes in the form of righteous incivility (RI), by which I mean incivility that is at least believed by proponents to be righteous, whether or not it is righteous in fact (defined subjectively enabling us to use the term without committing to the righteousness claim). WC and RI are the main candidates for justified VC. My focus, therefore, will be on the advantages of AC resistance over WC and RI.

WC and RI have something in common: they both involve harm. The harm done by WC is directly incurred by victims of wrongdoing, or those who get silenced or ignored for speaking up on their behalf. RI, by contrast, directly harms wrongdoers. Both approaches assume that one must either permit wrongdoing or attack wrongdoers. But AC resistance maintains that this is a false dichotomy. It is not a zero-sum game. In fact, as we shall see, it is precisely the shared element of harm that causes both WC and RI to backfire, rendering them ineffective. As the Hegelian approach to conflict predicts, by retaining the virtues of WC and RI (their narrow ends) while subtracting their shared vice (violent means), AC resistance achieves the synthesis that improves upon both: the more effective promotion of moral ends through moral means.

AC > WC

Take WC first. Here, we can be brief, given my argument in part one. WC tries to avoid violent means. That is a virtue. Unfortunately, it often does so in a cheap way by focusing on means that are “nonviolent” in a watered-down sense such as politeness (or law-abidingness, etc.). Civility in this sense often causes harm in several ways already discussed. Insofar as civility circumvents this, it does so by employing nonviolent strategies introduced in part two. Mere civility is not enough to achieve its goals; it needs nonviolence. AC resistance by contrast directly maintains that we must avoid harm as a means. Therefore, only when we adopt the appropriate means do we see the prospect for success, as the FHGM predicts.

Interlude: Engaging RI with AC

Like RI, AC allows us to resist injustice actively and forcefully. Like WC, AC opposes RI. But unlike WC, AC does not do so by ignoring or silencing the righteously uncivil. Instead, AC advocates must stick to AC in their opposition to RI. We have recourse to the methods introduced in part two.

First, we can use what I termed “ahimsic distraction”: we can join fans of RI in their fight against injustice, demonstrating the power of nonviolence in the process, hoping to inspire fans of RI to adopt our method. We can also use what I called “positive ahimsic force”: calling people up (rather than calling them out) via powerful speech that appeals deeply to their conscience.

Of course, many fans of RI will not be convinced absent direct dialogue. Due to practical constraints, we might lack the time. If we must choose between addressing an injustice or directing our energies against those fighting it poorly, the nonviolentist focuses on addressing the root harm: the injustice itself.

But often, we can make time for dialogue, as effective resistance movements do. And if we can, we should, since dialoguing with our own side and advancing our cause go hand in hand: to address the injustice, we must transmit the methods that will advance the cause rather than stall it, set it off course, or undermine it.

How do advocates of AC dialogue with fans of RI? Instead of silencing or ignoring them, we listen to them, hear their pain, validate their concerns, empathize, learn from them, and care enough to work with them rather than against them to find a more effective method to resist injustice. In other words, the best tool for the job is Marshall Rosenberg’s Nonviolent Communication (NVC). NVC enables us to converse in more human terms, to connect at the level of feelings and needs, opening the door for fans of RI to hear the case for AC with the reciprocal willingness on our part to open-mindedly hear their case. Then enters ahimsic argumentation. As you can see, I am especially concerned to extend care to audiences that feel angered at the suggestion to avoid RI. In what follows, however, I assume an audience that welcomes argumentation on this subject.

AC > RI

The first step in the argument against RI is to acknowledge its many virtues. It enables us to resist injustice, speak truth to power, express solidarity with victims, raise consciousness, mobilize the complacent into action, and ultimately advocate for change. It also provides an emotional outlet for the very real pain behind it.

Notice, however, that these virtues do not entail harming the “opponent”—intended here in a morally neutral way to refer to those who disagree with one’s side. Again, it is a false dichotomy to assume we must either admit defeat or harm the other side. There is a third option, namely persuasion: to win over hearts and minds. To the extent that this is successful, it benefits victims by removing opposition.

I suspect that fans of RI discount persuasion primarily because they assume that in the kinds of cases in which we contemplate RI, the divides are too great. Winning opponents over is too utopian. However, this assumption is unwarranted.

First, persuasion comes in degrees, ranging from fully convincing wrongdoers to convincing them to at least be more tolerant or accommodating despite continued disagreement. We can often make progress in this direction even if we cannot fully win them over.

Second, given the right methods, full persuasion is possible no matter the breadth of the divide, as the examples of Black, Davis, Lawson, and Tuff demonstrate. AC methods work because they deal with the emotions of opponents, thereby avoiding the backfire effect of VC. Had Black’s Jewish classmates, Davis, Lawson or Tuff used RI, Black would have remained a white nationalist, there would be hundreds more KKK members in the world, the angry biker would have continued pummeling Lawson’s friends, and the gunman at Tuff’s school would have pulled the trigger.

Third, even when we cannot persuade, RI makes matters worse than they already are. It invites retaliation, whereas AC deescalates. RI also undermines our ability to combat misinformation—a root cause of violence. Social epistemologist C. Thi Nguyen distinguishes between epistemic bubbles and echo chambers. One is in an epistemic bubble when one fails to receive information that conflicts with one’s views, which is primarily due to social selection filters: we tend to associate with like-minded sources of information (families, social media friends, preferred news outlets, etc.). The solution here is relatively simple: pop the bubble by providing the information. We can do this by fact-checking, seeking viewpoint diversity by spending time among others with diverse viewpoints, and consuming a range of news sources—a practice philosopher Sanford Goldberg calls coverage-reliability.

One is in an echo chamber when one preemptively dismisses any conflicting information as unreliable because it comes from outsiders. The authorities trusted within an echo chamber often reinforce evidential preemption via conspiracy theories, fake news, fear-mongering, and hateful speech. In this case, fact-checking, evidential bombardment, and coverage-reliability do little good. This is why it is difficult to break people out of cults, including what Steven Hassan, one of the world’s leading cult experts, calls The Cult of Trump.

RI in the form of vitriol cannot break through the chamber walls. It instead invokes retaliation. Of course, vitriol is but one form of RI. RI can also come in the form of unfriending, canceling, or deplatforming. Some such distancing tactics may be justified in special circumstances, namely when they are necessary for self-protection. But they are forms of RI when wielded as punishment. Either way, distancing tactics might mitigate the backfire produced by vitriol. But they cannot dissolve barriers between groups. They instead strengthen existing barriers, ultimately reinforcing echo chambers as well as epistemic bubbles. This might be defended on the grounds that the thicker the barrier between ourselves and wrongdoers, the better we can moral grandstand or virtue signal to victims that we are on their side. Following Kenneth Burke in his classic book A Rhetoric of Motives, rhetoricians call this identification and division: the process of dividing oneself against some to identify with others. Burke maintains that these two processes are inextricable. But there is another option: to identify and persuade rather than divide. And again, by leaving a source of opposition intact, division harms victims.

Think about the ramifications of division as a general tactic for a world as interconnected and interdependent as ours. In 2024, Americans are evenly split between Trump supporters and people who would vote for nearly anyone else on earth for president. Many of the former live in blue states and many of the latter live in red states, not to mention the purple states. We are likely to share a workplace, living space, classroom, public venue, etc., with someone across the political aisle. Even if not, our “friends” will have “friends” with such connections (or the “friends” of our “friends” of our “friends” …). This follows from Stanley Milgram’s six degrees of separation, a version of the “small-world hypothesis.” Now add the fact that one’s political party is only one of many dividing lines: there are major disagreements among Republicans and among Democrats, and some of those disagreements crisscross party lines, including political, economic, ethical, religious, and cultural disagreements. Moreover, people sometimes switch parties, and others are apolitical, including the children of our “enemies.” We cannot feasibly cordon off one group from the other with respect to any single major division much less all of them. It would be like carving up society with a hacksaw! Yes, there are many on each side who hate the other. Even if you believe this justified, the practical truth of the matter is that we simply must figure out how to work with them anyway, however offensive or dangerous we find them. There simply is no other way for society to survive.

As Nguyen argues, there is only one way to break through the walls of an echo chamber. The problem is distrust. Clearly, then, the only solution is to gain trust, which requires demonstrating goodwill. We can do so using Rapoport’s rules, NVC, and the other methods explored in part two. These methods are specifically designed to work at the emotional level, reopening the rational channels through which information can flow. By getting interlocutors to see us as a trustworthy people of goodwill, they begin questioning their default assumptions, initiating what Nguyen calls a social epistemic reboot in which one reexamines from scratch which sources are reliable. It is a slow-going process. It requires patience and sustained dialogue. But this is to be expected when dealing with an entire worldview constructed over a lifetime. And it’s the only effective method we’ve got, even with respect to ends narrowly construed. This conclusion is reinforced by Hassan, a former cult member who has helped many others deprogram. Hassan finds that conversation demonstrating trust and goodwill is the only way to break people free from the grips of a cult.   

So far I have made three points in response to the objection that we cannot use AC to persuade those across the aisle of a deep divide. A fourth response is that the propensity to persuade opponents and avoid escalation are not the only strategic advantages. AC also has a greater propensity to persuade neutral parties. VC tends to backfire when it is noticeably used in an asymmetric fashion: upon witnessing one side use violence against those who remain nonviolent, neutral onlookers will tend to support the latter (moral jiu-jitsu). This effect is amplified by the strategic use of media. But the effect is undermined when both sides use VC. In that case, each side will cheer on their own VC while using the other side’s VC as evidence of hypocrisy, canceling each other out. Neutral parties typically find all of this off-putting and see no reason to budge. We see this recurring dynamic in presidential debates. AC resistance, however, is designed to avoid backlash by working at the emotional level to attract third-party support.

There is data on the effectiveness of violent versus nonviolent resistance movements. In their book Why Civil Resistance Works: The Strategic Logic of Nonviolent Conflict, Erica Chenoweth and Maria Stephan examine the data, finding that nonviolent resistance campaigns are over twice as likely to succeed, even when we limit attention to the most extreme cases (oppressed peoples under brutal dictatorships).

From “Confronting the Myth of the Rational Insurgent” by Erica Chenoweth
(reproduced in their book with Maria Stephan)

The data also revealed why: (a) movements with greater numbers of diverse participants are more likely to succeed, (b) movements are more likely to attract greater numbers of diverse participants by avoiding fear tactics and violence, making it feel safer to join, and (c) nonviolent tactics avoid such tactics, thereby attracting greater numbers of diverse participants. Chenoweth and Stephan were not looking at communication. However, the same basic factors plausibly apply: we get more people from diverse backgrounds to feel safe in joining the conversation and speaking up when our communication avoids vitriol. With people power on our side, we increase our chances of success.

The fifth and final response to skepticism about winning over opponents is that where we fail, there remain other nonviolent options. These include negotiation, compromise, protest, distraction, and noncommunicative forms of nonviolent action.

So far I have been exploring the contingent but strong causal connection between means and ends. While this will continue to play some role in what follows, I now turn to the constitutive connection, in particular its phenomenological aspect. This primarily pertains to the emotional content of communication. The emotion primarily associated with RI is righteous anger.

Righteous anger is often cited as the energy that motivates action. AC, one might object, loses this benefit, resulting in complacency and the perpetuation of harm. There are two lines of response. First, we might argue against anger. Martha Nussbaum makes this move in her book Anger and Forgiveness. Following Aristotle, Nussbaum analyzes anger as an emotion that includes a retributive instinct. Insofar as retributivism is unjustified (to be discussed below), this renders anger unjustified. Instead, Nussbaum advocates “transition anger,” an emotion that is something like “How outrageous! Something must be done about this.” This emotion isn’t quite anger because, although it wishes for something to be done, it doesn’t wish specifically for harm to befall the wrongdoer. In conjunction with compassion and empathy for victims and hope for perpetrators to become better, transition “anger” gives us all the emotions we need to motivate action without the problematic retributive component.

Alternatively, some nonviolentists accommodate righteous anger, suggesting that we learn from it and channel it in a more constructive direction, namely against deeds and systems rather than persons. See, for example, Arun Gandhi’s The Gift of Anger, and Rosenberg’s chapter on “Expressing Anger Fully.” Perhaps this ultimately amounts to the same thing as Nussbaum’s approach, the main difference being verbal/conceptual: the Aristotelian account doesn’t classify it as anger.

One thing incivility can do that AC cannot is fulfill the desire for retribution in response to wrongs: incivility is justified when deserved. Anger might be conceptually linked to retribution, but not vice versa: it is possible to have a purely rational belief in retribution, Kantian style. Either way, retribution does not help for 10 reasons.

1. Hegel was wrong that retribution “negates the negation” (71). Rather, Socrates was right that “two wrongs don’t make a right” (to use Benjamin Rush’s aphorism). Since harming the offender doesn’t return what was lost, it simply multiplies wrongs.
2. Retribution doesn’t address the underlying cause of the problem. Insofar as we don’t solve the problem, we permit further harm. This is one of the main points in favor of restorative justice.
3. Many cases of incivility do not harm their targets. For example, when media personality Tucker Carlson was called a “pig fornicator” on a private social media feed, it’s doubtful he heard about it much less cared (Keith and Danisch 147). If it didn’t harm him, it functions poorly as retribution.
4. Even when RI fails to harm targets, as in the Carlson example, it harms others by setting a bad example, whose incivility in turn perpetuates the other harms attached to RI.
5. RI harms oneself. This is partly due to my earlier point that RI invites retaliation, and therefore tends to escalate rather than deescalate. Moreover, as the Confucians argue, each act of incivility makes the next easier. In this way, one negatively shapes one’s own character. Relatedly, as the Buddhists argue, giving into anger, hatred, and other reactive emotions associated with RI undermines rational control, tending to blind us and promote haste. These points about self-harm explain the idea that “hate is a prison.”
6. Insofar as hate is a prison, it follows that, by giving in to the temptation for RI, we have thereby permitted the other side to imprison us. And that in turn means they have a kind of power over us. By our granting them that power, they have already achieved a victory of sorts, especially if their goal was for us to take the bait.
7. By resorting to RI, one has resorted to the very behavior one rejects. This is a double-standard. By responding to the other side with RI, one signals to them that it is permissible for them to do likewise. To be consistent, one must condone it for everyone or no one. Here one might appeal to the Kantian Formula of Universal Law: act only according to maxims that you can rationally will to become universal law. It is better to “rise above.”
8. Consider the special case where RI is based on falsehood (e.g., the Carlson example). When someone behaves badly, the badness can be captured by the truth about what they did. Anything beyond the truth exceeds what they deserve. What’s deserved cannot be based on falsehood. Falsehood therefore cannot be retributively justified.
9. In the special case where RI takes the form of ridiculing a wrongdoer for a morally neutral quality, we inadvertently ridicule others with the same morally neutral quality. For example, when Joe Biden is ridiculed by conservatives for his stutter, they inadvertently ridicule everyone with that speech impediment. When Trump is called “fat” by liberals and ridiculed for purportedly wearing a hairpiece, they inadvertently body-shame others for their weight and hair insecurity—people who clearly do not deserve it. (Consider whether you would issue such comments when a friend meeting the relevant physical description sits beside you. I hope the answer is no. But then it would be two-faced to do it in their absence.)
10. For all these reasons, retributive incivility fails long term. The thirst is unquenchable, burns us up, consumes us, leaving us exhausted and threatening collapse into a defeatist attitude. Ultimately, retribution is neither a constructive energy from which to act nor a healthy emotional outlet because it leaves one powerless and ultimately unsatisfied. To quote the “Twin Verses” of the Buddhist Dhammapada:

“He insulted me, he struck me, he cheated me, he robbed me”: those caught in resentful thoughts never find peace.

“He insulted me, he struck me, he cheated me, he robbed me”: those who give up resentful thoughts surely find peace.

For hatred does not cease by hatred at any time: hatred ceases by love. This is an unalterable law.

The Ahimsic Portrait of an “Opponent”

As we have seen, harmful means yield undesirable ends. This is exemplified by WC and RI. But if we are to let go of thinking in retributivist terms and the associated reactive emotions, we must understand the alternative. How else can we conceive of wrongdoers? How can we not be angry at them or hate them or think they deserve punishment, whether in the form of RI or otherwise?

The first thing to do is retire what Barry L. Gan calls the “myth of good guys and bad guys,” which goes hand in hand with retributive thinking. Once we label someone a “bad guy,” we view them as nothing but a “problem”—an I-It rather than I-Thou relationship, to use Martin Buber’s distinction. This dehumanization makes violence psychologically feasible.

As Gan argues, the myth is underwritten by a host of logical fallacies and cognitive biases. First, it is a false dichotomy. No one is purely good or purely bad. Everyone is a mix of both good and bad characteristics. Whether we place someone in the good guys box or bad guys box is a matter of which characteristics are salient. If, for example, we only know one bad thing about a stranger, we tend toward hasty generalization: they automatically go into the bad guys box. Often what is most salient is someone’s social identity (e.g., political affiliation, religion, race, gender, sexual orientation), in which case we place the person in the bad guys box due to a false stereotype about the group as a whole. Regardless of what counts as the in-group versus out-group, “we” always go in the good guys box and “they” always go in the bad guys box. Those we place in the bad guys box do the reverse. And of course we can’t all be right.

This does not mean we cannot recognize bad behavior stemming from our side. But when we do, we don’t typically kick them out of the good guys box. Instead, we tend to view them as basically good people who sometimes behave out of character due to situational pressures. But when the same bad behavior stems from the out-group, they’re just plain bad. This double-standard is a well-studied cognitive bias known as the fundamental attribution error.

This error highlights two approaches to explaining bad behavior: we can attribute it primarily to the agent’s character or to situational factors. According to Stanley Milgram’s landmark obedience study and the situationists who followed, wrongdoing is largely driven by the external situation one is in rather than internal character traits. This is how Milgram explained how many ordinary people in Germany cooperated with Nazi atrocities. On his view, most of us are capable of doing terrible things. It’s largely a matter of whether we find ourselves in a situation that activates that potential. Most of us vehemently deny this, but studies reveal that humans are terrible self-predictors with respect to novel situations. Most people surprise themselves.

The appeal to situationism does not undermine nonviolence. On a moderate interpretation of situationism, it is still possible to develop a robust nonviolent character, though it requires great effort, and most people are not there yet—including most advocates of nonviolence, as we freely and openly admit of ourselves. Insofar as we fall short, we admit it is a flaw, a shortcoming, something to overcome. On this understanding, wrongdoing is a commonplace error to correct, to be distinguished from blameworthiness (deserving of anger or other negative reactive attitudes or emotions). Rather than punish ourselves or others for the leaky boat we coinhabit, we replace negative reactive emotions with a constructive, problem-solving orientation, using the acknowledgment of error as motivation for continual improvement, inching ourselves and others ever closer to the goal of complete nonviolence.

The Buddhist analysis proceeds one step beyond situationism. Buddhist ethics emphasizes that one’s internal character and immediate external situation are but two elements within a complex web of interdependence. In the case of bad behavior, there is the deed itself, the choice to perform the deed, the person’s bodily condition, their internal psychological states, their underlying character, the external situational factors (environmental, social, and systemic) that activate particular character traits, and the causal etiologies thereof. This entire complex is the culprit. The “person” or their character is but one node in this vast web. To selectively focus on any single node as the locus of wrongdoing is arbitrary, driven by the logical fallacies and cognitive biases above, and ultimately counterproductive. Instead, it is better to see the full picture.

When we examine the full web of causes and conditions, we find ourselves there as well. All humans are in it together, and we are each responsible to some degree. And if we were to trade places in the web with anyone else—inheriting their education, socialization, ailments, opportunities in life or lack thereof, the entire history of their experience, their resulting psychology, current knowledge, current options, current pressures and stresses—we might very well find ourselves behaving exactly as they behaved. I am not endorsing determinism here. I am not denying free will. Nor am I rejecting good versus bad, right versus wrong. But I am suggesting that whenever we see wrongdoing, our default assumption should be that some unfortunate conditions led to it. “Hurt people hurt people,” as the saying goes. This realization in conjunction with the preceding points should cause us all to have a bit more humility about ourselves, caution in our judgments of others, and compassion for whatever unfortunate conditions led others into wrongdoing. As Thich Nhat Hanh put it, “When you understand, you cannot help but love” (24).

Relatedly, in his sermon  “Love Your Enemies,” King instructed us to love the person but hate the deed, to hate the system rather than the person caught up in the system. Two points of clarification. First, we should be wary of abusing this point. It applies only where the deed is truly wrong and the system is truly unjust. Second, loving someone does not mean permitting their bad behavior. As Gandhi notes, love requires resistance to wrongdoing. A parent who loves their child must resist the child’s wrongdoing, albeit with care, viewing them with compassion and wishing them to become better than they are. And if we can’t help the wrongdoer become better despite our best efforts, then in serious cases, in cases where they are especially dangerous to themselves and/or society, we can offer protection via “benevolent quarantine”—isolation or restraint to the minimal extent necessary while carefully attending to their well-being. This follows Gregg D. Caruso’s public-health quarantine model as an alternative to prisons and other forms of punishment. On this model, we view perpetual wrongdoing as an affliction to be healed. Following a recent innovation in trauma response, I suggest replacing the affliction-first language (“ ________ person”) used up to now with person-first language (“person experiencing ________”) to emphasize that there is more to the person than their afflictions.

When we view a pattern of wrongdoing as an affliction that calls for attentive care and protection, focusing on the web of causes and conditions that collectively underpin wrongdoing, destructive anger and blame tend to transform into compassion and empathy coupled with a constructive problem-solving mindset. Our attention will then be devoted to addressing the causes and conditions that stimulate violence in the first place. AC, and nonviolence generally, prompts us to address the underlying problem, or at least use techniques such as distraction to block escalation and provide protection.

Of course, it is difficult to transform our deeply entrenched habits. In the grips of intense emotion, we cannot simply shut it off. However, in the long term, there is much we can do to retrain our habits. Personal cultivation is at the heart of the Kingian and Gandhian traditions. Among his Six Steps for Nonviolence Social Change, King includes “Personal Preparation.” Before King, Gandhi emphasized “self-purification.” And for combating our proneness to anger and blame, there exist a wealth of practices from the world’s philosophical traditions ranging from Buddhism to Confucianism to Daoism to Stoicism to Quakerism. Some of these are anger-elimination techniques. Others focus on anger-redirection techniques. Whichever methods we adopt, they also apply to our attitudes toward those in the throes of anger and blame and their manifestations in VC. We must not blame those who blame or hate those who hate or get angry at the angry. We must proceed with care, using AC against VC.

Conclusion

Nonviolence is misunderstood by many opponents as inaction. It is also misunderstood by some proponents as a mere tactic to be employed when necessary. According to the portrait sketched above, nonviolence is more deeply a way of seeing, a way of feeling, and a way of being in the world. This way of seeing, feeling, and being does not remain internal and private. It manifests externally by driving our actions upon the world. Nonviolent action is immensely practical because it is moral. It tends to produce desirable ends due to moral means.

This does not mean that by using nonviolence, we always get everything we want. In that sense, sometimes nonviolence fails, sometimes violence succeeds, and sometimes no method prevails. But the tendency for nonviolence to yield better results remains. We cannot time travel to the future to determine what will succeed. We must make our judgments looking forward with limited knowledge. And nonviolence is our best bet. Even when it fails, we cannot judge decisions in retrospect. If it doesn’t rain despite the 89% chance, bringing the umbrella was still the right call.

This returns us to Gandhi’s point about control. Given that we cannot predict or control the future, doing so cannot be our obligation. To quote Holmes, “our obligation in particular situations is not to change the external world; it is rather, to try to do so, and to try as carefully and responsibly as possible; and to do so in ways that are morally best” (226). Failing to try guarantees failure. Trying gives us a fighting chance. And if we try but fail, nonviolence is the only guarantee that we at least do not add more harm to the world. In this way, when we fail, nonviolence gives us the assurance that we did it right. In that respect, it is up to us whether we succeed.

But nonviolence does not come naturally. It must be learned and practiced. One place to start is with our communication—one of the central ways we act upon the world. Mere civility isn’t enough. Righteous incivility is not the solution. My modest hope for this series, which now draws to a close, is to plant the seed for ahimsic communication. I hope you will join me in taking it up.

Acknowledgments:

Many thanks to Linda Mills Boyd, Alexis Flint, Barry Gan, Bob Holmes, Loren Manchester, and Jelena Markovic for their insightful feedback on earlier drafts of this piece.