When Does “Being Rational” Lead You Astray?
The Professor Who Just Couldn’t Be Bothered
Imagine a professor—let’s call him Procrastinate. A publisher asks him to write a book review. He knows he ought to do it, and he even intends to. But here’s the catch: the first step is simply to accept the job. And even if he accepts, he’s almost certain he won’t actually finish. So, should he accept? It seems like saying yes would be pointless—or even a mistake. Yet if he doesn’t accept, he’s ignoring a necessary step toward a goal he genuinely thinks he should achieve. That’s a strange place to be. Philosophers use puzzles like this to pull apart two ideas that often get tangled up: being rational and having good reason.
Two Kinds of “Ought”
When someone asks, “What should I do?”, they might really be asking two different questions. One question is: “What would make my thoughts and plans fit together neatly?” That’s about rational coherence—having your intentions, beliefs, and actions line up without clashes. The other question is: “What do I actually have a reason to do, all things considered?” A reason is a fact that counts in favor of doing something, even if it doesn’t force your hand. For example, you might have a reason to eat cake (it tastes good), but stronger reasons to eat a salad (your health) outweigh it. So you ought to eat the salad, even though the cake reason still exists.
Sometimes these two questions point the same way. But often they can pull apart. Suppose your strongest desire is to stay healthy, but out of habit you still intend to light a cigarette. If you then also intend to smoke, your plans are locally coherent—the intention to smoke follows from your belief that lighting up is the means. Yet you don’t have good reason to smoke, because that desire is actually undermining your deeper desire for health. So you’re being rationally coherent with a bad plan.
Things get even messier when we consider that what you have reason to do might depend not on what you think is true, but on what’s really true. Imagine someone whose deepest desire is a gin and tonic, and she believes the clear liquid in a bottle is gin. It’s actually petrol. According to a view that ties reasons to objective facts (the Objective Desire-Based Theory), she has no reason to mix that liquid with tonic. Yet if she forms the intention to do so, her attitudes are perfectly coherent. The same split shows up if we reject desire-based views entirely and adopt a Value-Based Theory: what you ought to do depends on what’s genuinely valuable, not just on what you happen to want. So a madman whose strongest desire is to start a nuclear war might be instrumentally coherent—his means line up with his end—but that doesn’t mean he has any reason to push the button. Philosophers like David Hume (1711–1776) and Immanuel Kant (1724–1804) took very different stands on which side of this divide matters most, and we’ll return to them.
When Means Get Messy
Philosophers often say that reasons “transmit” from ends to means. If you have a reason to achieve something, then—because of that—you have a reason to take the steps that help you get there. But the details are thorny. Consider a doctor whose patient is in pain. Drug 1 is a sufficient means to relieve the pain; it works perfectly on its own. Drug 2, if given together with Drug 1, first neutralizes it, then recombines to produce a third drug that also relieves pain. The doctor is already going to give Drug 1. Does she have any reason to give Drug 2 as well? It seems not—even if Drug 2 costs almost nothing, using it feels pointless. That’s because Drug 2 is superfluous: the end would be achieved regardless. So a simple rule like “If you have reason to achieve an end, then you have reason to take any sufficient means” runs into trouble.
Other means are costly or outright objectionable. Flying to Alaska might be one way to reach a new job, but if the ticket costs more than you’d earn, you probably shouldn’t fly—and maybe you don’t even have a reason to. More dramatically, suppose the only way to improve your school’s unfair policy is to harm an innocent person who blocks every change. Many people would say you have no reason to harm them, even if doing so is a necessary means to a worthy end. The link between ends and means isn’t automatic: it can be broken when the means are too awful, too expensive, or just not needed.
And then there’s Professor Procrastinate again. Accepting the commission is a necessary means to writing the review. But since he’s so unlikely to finish, does he have the same strength of reason to accept as someone who is reliable? Probably not. All these puzzles show that the simple idea “if you want the end, you should want the means” hides a landscape of tricky exceptions.
Why Your Brain Hates a Loose End
Despite all the complications, there’s one thing most philosophers agree on: if you intend to do something, believe that some action is a necessary means to it, and yet refuse to intend that action, you’ve violated a kind of standard. Call this the Violation Claim. For example, I intend to arrive at the party on time, I believe I can only do so if I intend to steal my neighbor’s car, but I refuse to form that intention. Something has gone wrong in my head—I’m being instrumentally incoherent.
But what exactly is this requirement I’ve broken? One possibility is that it’s a genuine requirement of reason: I ought to be instrumentally coherent, just as I ought not to cause pointless suffering. The worry, though, is that this leads to “bootstrapping.” If the requirement says, “If you intend an end, you are required to intend the means,” then, when I intend to do something awful—like stay in power by killing a rival—I would be required to intend the killing. That seems crazy. So many philosophers suggest the requirement takes a wide-scope form instead: you are required to either give up the end, or revise your belief about the means, or intend the means. That way, no one is forced to intend anything horrible; they can just drop the bad end.
The wide-scope view has its own problems. If I ought to mail a letter, then—by similar reasoning—I ought to either mail it or burn it. If I burn it, I’ve done something I ought to do according to that strange disjunction, but that’s clearly not what we mean. Defenders of the wide-scope approach have to explain why satisfying the disjunctive “ought” still counts as satisfying a rational requirement. The debate remains lively.
Maybe It’s Not About Rules at All
Some philosophers take a step back and ask: when we call someone “irrational” for being instrumentally incoherent, are we really pointing to a rule they broke, or just describing how they’re functioning? One view says that being instrumentally incoherent is a sign that you’re not responding properly to your own beliefs about what you have reason to do. In other words, the wrongness is just that you’re ignoring your own internal compass. When you become coherent by dropping a foolish end, you’ve satisfied a standard—but that standard might not be a fact about what you ought to do; it might only be a fact about what you already think you ought to do. The requirement could be merely apparent normativity.
Another possibility is that rational requirements are just standards of proper functioning, like a well-working heart or a reliable bicycle. A heart that misses a beat isn’t wrong in the way a lie is wrong; it’s just not operating as it should. Similarly, an instrumentally incoherent person isn’t necessarily violating a rule of reason—they’re just not functioning perfectly as an agent. David Hume himself took a skeptical line: because intentions and desires aren’t true or false, he thought they couldn’t be called “unreasonable” in a strict sense. On his picture, only false beliefs could be blamed for mistakes, not the plans we build on them. Many later philosophers have pushed back, arguing that we clearly feel a special pressure to be instrumentally coherent—a pressure that needs explaining, not dismissing.
Why This All Matters When Your Plans Clash
You don’t have to be a madman or a procrastinating professor to feel the pull of these puzzles. Think about a time you wanted to succeed at a sport, but the only way to practice enough was to skip homework and let your grades slip. Your desire to excel in one area bumped against your desire to do well overall. You might have felt that sticking to the sport plan was “rational” in a narrow sense—it made your practice intentions coherent—while still sensing you had better reason to prioritize school. The friction between being coherent and doing what’s genuinely best is a daily human experience.
Philosophers call this tension the difference between instrumental rationality and substantive reason. Instrumental rationality cares about the shape of your plans; it’s like the grammar of your decisions. But just as a sentence can be grammatically perfect and still false, a plan can be instrumentally flawless and still a terrible idea. Understanding this split doesn’t tell you what to do—but it does give you a sharper tool for thinking. When you feel stuck, ask yourself: am I trying too hard just to make my intentions fit together, without stepping back to check whether my whole aim is worth keeping? Both perspectives matter. And even if philosophers still argue about which one is more fundamental, you can use the distinction to untangle your own choices—one loose end at a time.
Think about it
- Suppose you really want to win a video game tournament, but the only time you can practice is late at night, and that would leave you exhausted and grumpy with your friends. Is it still rational to practice? Is it the right thing to do?
- Imagine you set a goal to become popular, and you believe the only way is to spread a mean rumor. If you refuse to do that, are you being irrational? Why or why not?
- Can a person make completely coherent plans—with every step neatly lined up—and yet still be a bad person? What does that tell us about the difference between being rational and being good?
