My grandson, according to my kid, really hates being wrong. He’s six, so I’m hoping that he lives long enough to know that being wrong is not only normal but often funny. I’m working on maintaining that attitude myself.

This morning, with my handy italic fountain pen, I wrote today’s date at the top of the page in black letter Gothic, one beautiful character at a time, with the ink flowing smoothly onto the page of my journal, thusly:

Tuesday, April 2026, 28.

In the case of the date this morning, it’s clearly not my fault but the fault of the numbers, what with all the 2s, because once I wrote 2 after April, my meticulous pen just marched on into 2026, the hell with it, and I didn’t notice until I had written the 6 with a nice little flourish that I had messed up, and I hastened to add the 28 onto the end. I left it the way that it was, because it was amusing, and because it doesn’t matter.

According to my journal, now that I look back, there were apparently two Fridays in the past week, too. I even thought once, as I was making the “F,” how funny it was that the previous page had an “F” on it too.

A couple of months ago, my adult kid called me upset with themself because they had made some basic mistakes about something, and I sent them snapshots of all the recent times I carefully wrote the date completely incorrectly in my journal. It cheered my kid up considerably. That kind of thing makes up most of the good advice I give to people, actually. They call me, shattered and dismayed, and I sympathize and tell them about all the times I have failed worse, and then I tell them they will be okay. It seems to cheer them up. Hell if I know whether it works or not, it’s all I got.

Because failure is everywhere. Error is normal, and it’s baked into the universe.

Alexander Pope wrote in his An Essay on Criticism in the 18th century, “To err is human,” but I think that’s not a big enough statement. No, a steady, annoying wrongness is part of the universe. Entropy, which is to say disorder, grows implacably, whether humans are involved or not (though we do slow the process down temporarily).

And all my life, I have watched as we humans tried to prevent error by improving technology.

I learned to touch-type on a big old Remington manual typewriter, in a summer course at a junior college when I was entering 10th grade. Back then, we were taught to sit bolt upright when we typed, and hold our hands just so in a T. rex position. It took a lot of strength and good posture to depress the keys with sufficient force while not messing up the back. The instructor played music at a metronomic rhythm, to drill us into a steady pace.

When I got within five characters of the right margin on that Remington, a bell would sound. The carriage return was a big lever, and the carriage really did return when you pushed the lever, stopping at the left margin with a thump that shook the table. If I made an error, I had to use a typewriter eraser to scrub the error off the page, brush away the resulting crumbs, and then carefully position the typewriter so that my correction landed in the right spot on the paper. If you erased too much, you made a hole and had to start over, or else patch the page with glue and a scrap of paper.

I became a very good typist, but I still always discovered errors after I rolled the paper off the platen, when it was too late, so my corrections would end up off-kilter and not quite aligned.

I took a portable typewriter to college, even though I still wrote many assignments in pen, on lined paper. Later, I graduated to an IBM Selectric with its whirling type-ball and unmoving body, and the sticky tape that allowed you to lift off your errors without having to position it. After the Selectric, I used a miraculous one-line word processor (in the late 70s) and then (in a law firm) an absolute marvel of a computer with a 20-meg hard drive.

We had an entire proofreading department at that last job, because it was a multinational firm, and errors could make the difference between winning and losing. Errors slipped through, no matter what we did.

Autocorrect appeared a good bit later on, and of course, that was miraculous. However, as a result, I could make more errors with fewer consequences, and so I did.

Now, when I type, I sit casually at a keyboard, my fingers fluttering almost silently, proceeding forwards and backwards, moving, replacing, and deleting letters, words, and paragraphs. My computer detects typos beautifully, and often suggests the correct word or phrase before I get to it–this line alone had several suggestions and a number of automatic corrections. I’m an even faster typist as a result, but a good third of my progress is backwards, because I keep changing my mind.

But autocorrect is also often just plain bad at its job, and even though technology improves in that area, it seems to magnify the mistakes it doesn’t catch, or the errors the technology makes all on its own. AI operates on rules of statistical probability and not thought, so it can make an absolute hash of wrongness look plausible. It is a copious manufacturer of predictable text. I have AI turned off everywhere right now, though I don’t know how long I will be able to keep my finger in that particular dike.

I’m a proficient proofreader; a lifetime of voracious reading and writing means I see the shape of words, not just a sequence of letters, and can pick out a typo from just glancing at a page and noticing where my eye catches on something. The problem is, when AI makes a whole sentence, paragraph, or page fictitious but plausible, I have to slow down too much, and read it word for word, letter for letter, with a ruler under each line, and pay attention to the meaning as well as to the words.

And yet, as I demonstrated this morning, when I slow down, I still make typos, even in ink, in black-letter Gothic, on lined paper, with my complete attention.

To err is human, then, but being human also means a constant struggle against increasing disorder in the universe, and hopefully meanwhile holding on to a sense of humor.