My recent post on what the ancients knew about posting well borrowed an early Buddhist test for speech, from a an ancient dialogue called the Abhayarājakumārasutta, in which we're told that speech worth saying is true, beneficial, and offered at the right time. I closed the post with a version of the test adapted for social media: four questions to ask before posting—is it true, is it beneficial or harmful, is now the moment, am I saying this for the listener or for myself.
In a reply on Bluesky, raised two thoughtful questions/pushbacks. The first is what "beneficial or harmful" means in a friend-enemy political environment, where, for those embedded in it, harming the other side reads as a benefit to one's own. The second is the place of the audience in the test. The audience point I'll take up briefly at the end.
A reasonable set of questions, but the second is ambiguous: beneficial to whom? Harmful to whom? The world of friend-enemy politics we swim in today, especially online, makes doing harm to apparent 'enemies' a positive good, a benefit, for 'our side.' www.aaronrosspowell.com/3mlerwhwebs2...
Elliott is right that "is it beneficial or harmful?" is ambiguous in ordinary political vocabulary. In that vocabulary, harming an enemy is a benefit, at least to the speaker's side. But my framework was Buddhist, and Buddhists think about harm in a pretty specific way.
To harm someone is to inflict suffering on them and to do so out of greed, hatred, or delusion—the three roots Pali Buddhism calls unwholesome. The suffering itself and the mental state and motivations of the person causing it go together in the analysis. The upshot is that, even toward enemies, it's wrong and, well, harmful to cause harm. What we actually want is for our enemies to be less harmful, which means released from the unwholesome states that drive their harm-doing in the first place, and so less likely to inflict suffering on others, including on us. The metta tradition makes this explicit. The wish extended to enemies isn't that they suffer, but that they be free from suffering and from the causes of suffering. Which is to say free from the very dispositions that make them dangerous. The goal is not to balance the ledger of suffering across teams. It's to reduce the total, and to release everyone, enemies included, from the states that produce it. Inflicting suffering on the enemy doesn't subtract from the harm they cause. It adds to it. I've argued before for a broader politics of harmlessness. The orientation that takes harm seriously as a category isn't tribal in structure, and seeing that exposes much of what passes for ordinary political behavior as harm-doing in disguise.
The framework's commitment here is more demanding than it sounds. The clearest statement of it is in the Kakacūpama Sutta, the Simile of the Saw, where the Buddha tells his monks that even if bandits were carving them up limb from limb with a two-handled saw, anyone whose mind gave rise to hatred toward those bandits would not be following his teaching. The instruction is delivered to monks and is technically about monastic mental cultivation, but the principle of non-hatred it expresses runs throughout the canon. The bar sits deliberately at the most extreme imaginable case. If the rule of non-harm doesn't extend to the people actively destroying you, it isn't a rule. It's a preference. The framework's answer to "beneficial to whom" works the same way. Beneficial in the sense of reducing suffering, for the spoken-about, including those one would call enemies.
There are two reasons this matters, even setting aside the welfare of the enemy.
The first is that intentionally inflicting suffering costs the speaker something. To speak with intent to harm is to act from hatred. To make it a regular practice is to cultivate hatred as a disposition. There's a passage in the Saṁyutta Nikāya about a man named Bhāradvāja the Abusive, who comes to the Buddha and insults him at length. The Buddha asks Bhāradvāja whether, when he offers a meal to a guest and the guest declines to eat, the food remains with the host. Yes, says Bhāradvāja, of course. Then so it is with insults, the Buddha replies. Unaccepted, they return to the one who offered them. The parable cuts at one of the assumptions baked into the dunk economy, that the satisfaction comes from the impact on the target. The Buddha's answer is that the abuse stays with the abuser whether or not it lands. The broader Buddhist analysis pushes further. What stays isn't just the abuse but the disposition to abuse, practiced and refined and made habitual through repetition.
Online life, with its endless small occasions for contempt, is an efficient training ground for exactly the kind of disposition the framework is trying to dismantle. Every dunk reps the habitual muscle.
The second reason is pragmatic. Cruelty toward political opponents doesn't work as persuasion. Pile-ons don't move targets toward better positions. They harden defenses and confirm targets in their sense that the opposition is bad-faith. I've written before about how the proper goal of political speech is to win opponents over rather than to defeat them, and how the dunk-and-destroy mode works actively against that goal. The Dhammapada states the underlying point directly: "hatred is never appeased by hatred in this world." We are running an extended live experiment in the claim. The result so far isn't the far-right collapsing under accumulated dunks. It's escalation.
What Buddhism cares about most fundamentally in ethics—what runs underneath the criteria for harm and benefit and the rest—is intent. The state of mind from which an action arises. The four-question test in the previous post isn't just a procedure for generating correct outputs. It's a practice for becoming aware of the state of mind you're acting from. The point of "Is it beneficial or harmful?" isn't that the words yield a clean decision rule. It's that asking the question slows you down enough to notice what motive you're acting from. I've described this elsewhere as the discipline of noticing ill-will and resentment as they arise, before we act on them. Most of us, most of the time, post without ever consulting our motives. The test interrupts that. Once we consult honestly, many posts fail. Not because the criteria are exotic, but because we're acting from hatred or vanity or the desire to dominate.
An obvious objection: strict rules typically have exceptions. Surely sometimes harsh speech, even speech that produces suffering, is warranted.
Maybe. But a focus on exceptions tends to do two bad things. First, it leads us to find exceptions everywhere. We justify whatever we wanted to do anyway by framing the situation as exceptional. The person we're attacking, this time, really deserves it. The post we're about to send, this time, really clears the bar. The exception swallows the rule, and the practice the rule was meant to instill never gets built. This is the same trap I've described before in the context of being an asshole on principle: "You can't say, 'I'm moral, except for here,' because that 'except' means you're not moral in any broad sense." A practice of motive-awareness that builds in exceptions for the cases when your motives turn out badly isn't really an awareness practice. It's just a way of lying to yourself that you have one.
Second, the exception focus changes what the framework is for. The Buddha's account of right speech isn't an optimization function. It's a practice that builds awareness of our own motives and changes our habits of mind and action. Worrying about exceptions before doing the awareness work gets the order wrong. The exceptions question can come up. It just has to come after we've done the practice of asking, honestly, whether our typical posting clears the bar. For most of us, it doesn't.
On Elliott's second point, the place of the audience in the test, he's right, and the amendment he proposes is sharper than what I had.
The body of the original post drew the distinction between speech offered to a person and speech offered to an audience at length. The four-question summary compressed it out. Elliott's tripartite version of the fourth question—am I saying this for the target of the speech, for a like-minded audience, or for my own edification?—does the work my binary didn't. Take it.
The framework asks something demanding. It asks us to examine our motives before we speak, and not flinch when the answer is unflattering. Most of us don't. Most online discourse shows it. The pushback that the test is too rigid for our political moment is often a version of the same avoidance, looking for reasons not to do the work. The work is hard. But it's also the point.