Magic on television is back in a huge way, and Netflix recently put up Justin Willman’s series “Magic For Humans”. Now, before I go any further here, I’m super worried that people will read the title, see Justin’s name in the first paragraph, and assume I’m calling him out for cheating. I’m totally not doing that.

The issue, though, is that he’s getting accused of exactly that. This video on tricking a spectator into thinking they’ve disappeared has been flooded with brutal comments, pointing out tons of details that make it seem obvious that the entire thing was staged. Strangely enough, the likes/dislikes ratio doesn’t reflect this attitude at all, with about 93% approval. Essentially, what it looks like is that we’ve got an otherwise popular video that’s found itself in the middle of a controversy.

What’s actually going on?

To answer that, we need to look into the weird domain of dual reality. Very quickly, dual reality is where different members of the audience are perceiving different things. This can lead to baffling magic for the greater audience, but it requires very careful management of all parties involved so that the principle isn’t tipped.

One of the real problems with dual reality is that while the audience has their minds blown, the assistant is usually hip to the fact that something more is going on, and their lack of reaction to what should be very strong moments of the trick can make things read false.

A crude example: The magician has a bag of random coins, and the spectator pulls one out. The magician correctly guesses the type of coin, which gets applause. The magician then asks them to picture a random animal. They do. The magician guesses “Duck?” and the spectator says this is correct.

Here’s where the dual reality part comes in. To the audience, they randomly selected a coin, which the magician guessed, and then the spectator randomly thought of an animal, which the magician also guessed. Assuming competent execution, the first part will be good, but the second part will be very strong.

In this situation, though, it’s not quite as strong for the spectator, because the coin that they pulled out also has “Duck” written on it with permanent marker. The script is just ambiguous enough that the spectator is induced to think of the written animal rather than a truly random one, but to the audience who thinks they’re all normal coins, it might as well be a truly random animal.

So, the main issue is that the audience is going to see two effects, but without extra preparation, the spectator is really only going to see one. There would probably need to be some added priming so that the spectator knows ahead of time that all the coins have animals written on them, and also knows that coins of any given type all have different animals on them (so that even though the magician correctly guessed that they were holding a quarter, for instance, there were still different animals on different quarters). Finally, in the same way that the audience got ambiguous wording and didn’t know that the coins have words on them, the spectator has to get ambiguous wording so that they believe that the audience knows about the words. It’s a very tricky process.

What does this have to do with televised magic? Well, you always want to try to get legitimate, credible responses from spectators. The problem is that you might have to mess around with the situation in a similar way to how dual reality gets employed. A long time ago, Steve Banachek did an AMA over on a old magic forum, and he was asked about the tactics Criss Angel used in his specials. At the time, consensus on the forum (if not the magic community in general) was that Angel constantly used stooges, and that any reactions elicited from people there were faked. Banachek offered a hypothetical scenario that would explain how an effect that seemingly must have been done with a stooge could actually be done “fairly”.

It would go something like this. Imagine that somebody is chosen to be on television with Angel, and the producer says that part of the deal is that there must be photos done. While getting ready to take the photos, somebody in the crew notes that the spectator is wearing a hat that’s branded Nike, and Angel has an exclusive deal with Adidas. Would the spectator mind wearing a different hat just for the shoot? The spectator obliges.

Later, when it comes time to do the trick, the spectator’s signed card ends up in their hat. To the spectator, this is an awesome trick because it’s their signed card. To the audience, this is an absolute miracle because the magician never had to touch the hat. All that’s needed at this point is to make sure that the audience never sees the earlier moment when the hats are exchanged, and you get an effect that couldn’t possibly be done without stooges, but with somebody who can truthfully swear up and down that they had nothing to do with it.

There are two major risks in this hypothetical scenario. The first risk is that the spectator says something to somebody about having to switch hats prior to the effect. I’m not going to go too in-depth on how the production crew would go about dealing with that, because everything I’ve written above is a pure hypothetical, and consistent only in spirit with the answer Banachek gave on the forum. The second risk is that the effect comes off as being too strong to possibly be legitimate, and as such must involve a stooge. This is a significant problem, since the more the person protests, the more people might dig their heels in and say the person is lying. If the Justin Willman video is any question, people might even research the situation and look into the assisting subject’s past, which raises creepy questions of internet vigilantism.

The criticisms of Willman’s video seem fair. People are complaining that the spectator sounds like they’ve been given a mic, which must mean they’re in on it. Truth be told, what might have happened is that the production team could have induced somebody to go watch a short magic show nearby, and then put a mic on him with the understanding that everybody else in the crowd was also mic’ed up. A similar protest comes up later when the television crew was filming the invisible guy lifting a wine bottle from the bag — if he wasn’t in on it, wouldn’t he find it strange that the cameraman could see him while the others couldn’t? Well, what might have happened is that after the initial “disappearance”, after Willman takes the call and steps away, the camera man goes up to the chair and asks “Hey, are you there? We’ve got some cameras around filming the people watching. If you’re here and you can hear me, go up to the bag over there and pull out the wine bottle. It’ll freak them out to see it floating.” The illusion to the spectator that people can’t see him is maintained, and yet there’s a credible reason for a camera crew being at the right place and the right time that doesn’t involve him being in on it.

Another example of this sort of possible finessing going on would be with the video with the kids who are left alone with a marshmallow, where they somehow make it vanish on their own. Again, notice the assumptions that we make. As an example of one direction this could go in, consider that we primarily assume that the kid is alone in the room. We don’t know for sure that there wasn’t somebody else there out-of-frame during the filming, and in between the shots, having some dialogue back-and-forth with the kid that would allow for a switch of a gimmicked marshmallow or gimmicked cup or whatever’s needed. This is where the dual reality might come into play again, with the kid impressed because they don’t know how the trick was done, but with the audience bewildered because everything that made it onto the final cut says that it couldn’t possibly be done.

Now, I’m not going to say that this is what happened. I have honestly no idea. He might have even cheated and used stooges. That said, I’m willing to take him at his word that he didn’t since the videos are almost certainly reproducible, right down to truthful reactions from the assisting spectators.

Regardless, it does create an issue with the television audience feeling cheated, because in order to prove that the spectators weren’t in on it, the producers would probably have to disclose information that might weaken the effect in other ways. There’s no way the marshmallow trick plays as well to a television audience if it’s known that the kid isn’t alone in the room, or that the “invisible assistant” plays as well if we suspect that the person didn’t just show up randomly. But, in an almost perverse twisting of the Too Perfect theory, leaving these things as they are can allow for suspicions of stooges get elevated to the point of conspiracy theory, and I don’t envy the position guys like Justin Willman are in because of it. Also, I’m probably waaaay overstepping my bounds by talking about this, since it’s not impossible that a google search might bring people here who aren’t magicians, but in a way maybe that’s ok, since it’ll give them something to think about before they automatically suspect stooges.