I’m one of those 40 million Americans who struggle with insomnia. And so I’m here today with an insomniac question: Why can falling asleep be so difficult?

There has been an exciting surge of research about sleep over the past twenty years. It is described very clearly by Matthew Walker—professor of neuroscience and psychology at UC Berkeley—in his 2017 book Why We Sleep. He identifies a variety of culprits that are now known to contribute to rampant sleep deprivation in the U.S. among insomniacs and others. They include: light from personal electronics; rigid and demanding work and school schedules; alcohol consumption; and the misconception that sleeping pills induce natural sleep.

All of this data is illuminating, and Walker’s book is well worth the read. But here I’m not looking for a third-personal story of the causal factors that adversely affect sleep. I’m asking a slightly different question: what explains that truly infuriating first-personal experience of trying desperately and yet failing miserably to fall asleep?

What explains this, I venture, is that falling asleep is not something you do.

What do I mean by that? I mean that falling asleep is not a mental action.

A mental action is something you do in thought. Some examples of mental action are: counting to ten; dividing 36 by three; imagining something to eat; rehearsing your lines before a play. Even though you can do all these things without moving parts of your body—as in bodily action—they are still things you do.

There are many different ways to sort mental actions into philosophically interesting categories. What’s relevant here is just one distinction between some mental actions and others. Sometimes when you perform a mental action, you determine exactly what you’re going to be thinking ahead of time. This is the case when you count to ten.

In other cases of mental action, you don’t pick precisely what you’ll think ahead of time. Performing this second kind of mental actions involves letting yourself be guided towards a determinate outcome that meets some target description.

Here’s an example: recalling when you last spoke with your mother. You can choose to do that, in those very terms. But when you set out to recall when you last spoke with your mother, you haven’t already picked exactly what you’ll think in the end (e.g. by saying to yourself “oh, I’ll just randomly choose to believe that I spoke with her last Thursday”). If you had picked that, you wouldn’t really be recalling something at all; you’d just be making something up. In order to count as figuring something out—which involves going after the truth—you have to let yourself be led by various facts about the past.

So in some cases of mental action, you have a complete conception of what you’ll think to yourself ahead of time, because you’ve already chosen it—as in the case of counting to ten. In other cases of mental action, you know what kind of thing you’re after in thought—e.g. the truth about when you last spoke to your mother—but you don’t pick ahead of time what you will end up thinking.

Which kind of thing is falling asleep?

I think it is neither. It’s not a mental action at all. It’s not even the kind of mental action where you try to do something by letting yourself be guided towards some specific type of thought. The outcome needed here is not some specific kind of thought (e.g. a recollection of when you last spoke to your mother last) but a total shutdown of mental action altogether, a complete surrender to a bodily process.

This is just a hypothesis, but I think it’s a promising one which can help us understand why it can feel so difficult to fall asleep. If falling asleep involves doing nothing mentally, then falling asleep is not just difficult to do; it’s something that you categorically cannot do. It’s not something you do at all.

Is there any sense at all, then, in trying to fall asleep? If my hypothesis is correct, then trying to fall asleep looks utterly self-defeating. It involves trying to perform no mental actions at all. But trying to perform no mental actions at all is itself something you do mentally: a mental action!

However, there is some subtlety here. You can try to fall asleep in the same sense in which you can try to sneeze on command. This involves putting yourself in a position in which an involuntary process takes over. To help sleep take over, you can take a hot bath; you can put on some white noise; you can (it is said) drink some warm milk.

You can even—surprisingly—try to perform some mental actions in order to fall asleep. The catch is that you have to do something so boring and easy that you’ll eventually stop trying and let some automatic process take over the task for you.

A canonical example is counting sheep. If what I say about falling asleep is right, then your counting sheep on purpose is not what lets you fall asleep; it’s your tendency to giving the task over to a mindless routine that lets you fall asleep. The counting of sheep just crowds out other mental actions that you are less likely to give over to some such automatic process.

Here’s the suggestion, in short: falling asleep isn’t something you do mentally, like solving a math problem or choosing a dish from a menu. On the contrary, what seems central to falling asleep is not performing any mental actions at all.

If my proposal is right, it could help explain the infuriating phenomenology of trying to fall asleep. Any attempt to fall asleep is necessarily oblique. You can put yourself in a position in which it’s likely that unconscious processes take over your stream of thought. But ultimately what matters to your success is that you stop trying to do things, and let such processes take over.

This hunch of mine follows in the footsteps of some other philosophical proposals about the relationship between mental action and sleep. Brian O’Shaughnessy, in his 2000 book Consciousness and the World, argued that dreaming involves a kind of mental paralysis which explains why our dreams make so little rational sense as they evolve. Matthew Soteriou picked up the discussion in his 2013 book The Mind’s Construction.

I’m not sure about this point, but I am sure there is an important (and still under-studied) connection between mental action and sleep. Philosophers must contribute expertise alongside scientific research to understand fully, from a first-personal perspective, why it feels so hard to fall asleep—at least for us insomniacs out there.