The serious work to be done in this essay is a consideration of what the word “only” contributes to the expression, “I was only joking.” This is something that might be said by someone who has just been reproved for passing along the scandalous “Pizza-gate” allegations about Hillary Rodham Clinton. In such a context, what function is performed by the modifier “only?”

But let’s begin with a lovely and completely fictional meadow.

There is a well-worm path in a meadow I am calling up in my mind. In the future, I suppose it will become a road and then maybe a turnpike, but for now it is just a footpath. But before it was a footpath, it was just an unusually flat section in the meadow—a sort of crease leading to the nearby woods.

Did I mention that I am making all this up?

And then, over the years, ninety-nine people walked along this soon-to-be-a path part of the meadow. The first 33 were headed to the woods to cut firewood and bring it back home by the armload. The second 33 were all a man named Per [1] who was sneaking off to meet Par [2], a beautiful Persian girl whom Per’s parents would like him to avoid. The third set of 33 trips were taken by a physicist deep in thought. She found the predictable walk to the woods and back…oh…restful. It cleared her mind to work on the space-time continuum.

OK, that’s just a little story I made up. Now, let’s get serious and talk about neurons. This picture represents a bunch of neurons that have never recorded any experience at all. They are as unspoiled as the never-before-walked-upon meadow.

OK, that’s just a little story I made up. Now, let’s get serious and talk about neurons. This picture represents a bunch of neurons that have never recorded any experience at all. They are as unspoiled as the never-before-walked-upon meadow.

Then it—your neural system— sees a letter H. And these 64 neurons ready themselves to fire. Sixteen are actually activated and the neurological infrastructure of your experience of the letter H is encoded there—but only for an instant.

And it leaves only this. The very faintest path, which encodes the memory of once having seen an H. Look again. Make sure you can see the faint trace.

The next time you see an H only a few of the neurons notice—let’s say the first one to notice was the one in column 3 and row 3, below—and that one triggers the activity in the rest of the trail. So the pattern of activation goes from the diagram in A to the diagram in B.

This storage scheme, the brainchild of psychologist Donald Hebb, is a powerhouse. Hebb proposed the mechanism a few years after World War II. Only within the past fifteen years, however, did researchers explore its mathematical premises and build large-scale computer models of Hebbian learning. Both endeavors—the mathematical insights and their implementation in computer simulations—have illuminated quite a few of the mysteries about why people think and feel the way they do. [4]

Every presentation of an H works the same way so far as these neurons are concerned and since they are the source of what you see, what they think is what matters for what you think. A prankster holding up a sign with an H on it, along with the text, “THIS IS NOT AN H” works. A legal citation to Section H, in HR 2243, works. A sign created to tell a school child what the eighth letter of the alphabet is, works.

The point? Everything that activates that path of neurons makes it stronger and more stable and more likely to override other patterns of neurons which might, in any given instance, be more nearly correct. The twentieth time you receive a stimulus, for instance, it might actually be an A. By that time, it doesn’t matter. You are primed to see H’s and that’s what you will see.

Does anything sound political yet? The neurological trace—now nearly a rut—winds up looking like this.

Solid looking, isn’t it? It is as solid-looking as an established footpath through a meadow. These neurons are like those blades of grass. The grass doesn’t care whether the shoes were being worn by the firewood carriers or the assignation keepers or the peripatetic theorizers. Every step that abrades the grass, thereby creating the path, makes makes every other user more likely to use the same path. And that is why it might be a turnpike one day.

During the 2016 election, I heard references to bizarre allegations that Hillary Clinton was running a child sex ring from a Washington area pizza parlor. I did not know until I read Benedict Carey’s piece]that those allegations were called “Pizza-gate.”

Here’s how that works. Some Hillary-hater devises the Pizza-gate scandal and posts it on Facebook. A thousand other Hillary-haters drink it down and believe every word. Eventually, one of that thousand, who has a Democratic friend or maybe just an academic friend, passes the Pizza-gate piece to someone who is outraged by it. This Hillary-lover passes the fake news story on to her friends along with a scathing commentary. Half of the people who get it from the Hillary lover think it is a spoof and one of the best “social media outrages” of the day and pass it along in a lighthearted way to other people who also don’t take social media seriously.

So here’s the thing. Every one of those people is walking along the same path. The dots go Hillary—Pizzagate—Sex scandal. And those neurons wake up the rest of the neurons in the chain. Connecting those neurons out of malice strengthens the connection. Connecting them as a fun media outrage strengthens them. Fulminating against them and demonstrating that they are false and malicious—STRENGTHENS THEM.

What to do? I really don’t know. I am distraught.

Refuting Falsehood

I’ve always been a fan of refutation. Refutation seems the best choice for falsehood. This has been said to be true, but it has been decisively refuted. We now know that it is not true.

My friend David Rawson and I once taught an interdisciplinary introduction to the social sciences. We arranged a model experiment for them. For our purposes, it was about where “men in general” carried their pocket handkerchiefs. To help engage the students, we agreed on a sampling rule for “men in general,” on on how the sample would be drawn and how big it had to be, and what level the findings had to be before either of us would be declared the winner.

So we ran the experiment exactly as we had planned it and one of us was declared the winner (I was) and the other hypothesis, which sounded entirely reasonable in the abstract, was declared to have been refuted. That’s what I like and what I am used to.

But Pizza-gate can’t be “refuted” if every attempt to spread it and every attempt to refute it work to strengthen the neurological connection. In fact, “refute” doesn’t really mean anything under those circumstances.

Shaming the Local Gossips

Back in the old days, the spread of information was slow enough and personal enough that lies could be nailed. “Oh, you got that from Harold? Then just ignore it. Harold makes up the truths he thinks will sell best.” Only the word “that” in that formulation refers at all to what Harold said. That story is now cast away and future stories made more doubtful because they came from Harold. Harold pays a price in this story for being the origin of or the purveyor of inaccuracies. If the price is high enough and he has to pay it often enough, he will stop if he is able.

Not any more

So back then, being the source of stories that turned out not to be true could really cost you something. Even passing along stories from notoriously unreliable sources could cost you something—but not as much. Now, by contrast, it costs the source of a defamatory and completely untrue story nothing at all. Whoever invented Pizza-gate, the story that Hillary was running a child sex ring from a Washington area pizza parlor, probably had a wonderful timing inventing it and posting it. It cost him nothing to do it.

If we were able to track down the person who did it, it would still cost nothing and if here were prosecuted, he would achieve hero status in the political tribe he belongs to. Needless to say, it cost the people who passed it along nothing. They could have stopped and checked to see if it were true—and in the small town of my example, someone might have—but stopping to check if something is true really does cost something. And no one expects a user of social media to stop and check whether a story is true before forwarding it to friends. Especially if you really hope it is true.

Wouldn’t it be just wonderful, this person might say, if Hillary were running a sex scam out of a pizza joint in Washington? I’m going to pass the story along to you so you can share with me the sheer joy of baseless malevolence. It costs me something to check on the factuality—and I don’t really care—and it costs me nothing to be found to be passing long false and malicious rumors.

The accusation of “only”

And that is what the “only” means in “I was only kidding.” I thought this was funny and I am passing it along to you because you will think it is funny too—and no consequences we care about will happen as a result of this “joke.”

And that might be true. Trust in the social institutions and political leaders that make a republic possible will be reduced. A reputation will be freighted with charges that do not pertain to her at all and that no one actually believes to be true. The insularity of the social network that passes these horrific stories around for the fun of it will be increased.

And yet, the accusation of the “only” in “only joking” might really be apt. Those really might be consequences you don’t care about.

[1] Short for Peregrine, it turns out. The traveler. So, technically speaking, these trips were peregrinations.
[2] Short for Parveneh. A Persian name, I was told by a friend who has that same name.
[3] This kind of activation and storage has been called Hebbian Learning, after Donald Hebb, in the 1940s. The instruments needed to verify it did not come along for decades but when the studies were done, they confirmed Hebbs’ theories.
[4] The diagrams and nearly all the analysis have been taken from a marvelous book called A General Theory of Love, by Thomas Lewis, Fari Amini, and Richard Lannon.