The original post for this episode can be found [here](https://johnaugust.com/2023/in-the-beginning-was-the-word).
**John August:** Hello and welcome. My name is John August.
**Craig Mazin:** My name is Craig Mazin.
**John:** This is Episode 607 of Scriptnotes, a podcast about screenwriting and things that are interesting to screenwriters. Today on the show, what are words, even?
**Craig:** What words are even is?
**John:** What words are even?
**Craig:** Is?
**John:** I promise I’m not high, Craig. What I really want to talk about is these fundamental units of writing and how weird words are, both for us humans and for computers. You’re a person who uses words a lot and who loves to play with them.
**Craig:** I love them.
**John:** Also, Drew has stocked up a lot of listener questions that we can go through. In our Bonus Segment for Premium members, Craig, let’s talk about Sinead O’Connor if we could.
**Craig:** We can. That will tie directly into my One Cool Thing this week, which is not just Sinead O’Connor in general, but a specific bit of Sinead O’Connor.
**John:** Fantastic. This episode you’re listening to was pre-recorded, but on Wednesday, tomorrow, you could be joining us live at the Dynasty Typewriter. We’re doing our first live show of this year, first live show in quite a long time. Tickets are all sold out, but you can still get livestream tickets. If you’re listening to this podcast right now, it’s like, “You know what? I really want to listen to John and Craig tomorrow and watch them with their special guests,” you still have the opportunity. We got a link in the show notes for that.
**Craig:** We’ve got some excellent guests. By this point, people know who they are.
**John:** By this point, people should. Tell us who our special guests are, Craig.
**Craig:** We have two currently nominated geniuses for two programs in the same category. Happily, I can report their friends. The great Natasha Lyonne and the great Quinta Brunson.
**John:** We’re so excited to see both of them and talk to them about writing in general and other fun stuff. They’re both great and geniuses and great performers. They’re going to be amazing guests. I’m really looking forward to tomorrow.
**Craig:** They are.
**John:** Before we started taping, we talked about extra special bits we’ll do. It’s going to be a fun show.
**Craig:** Now, John, what are the odds that the Writers Guild and SAG are going to picketing our live show? I don’t want to cross a line. Are we going to have to cross a line?
**John:** There will be no lines crossed. The only crossed will be lines of taste and discretion.
**Craig:** Oh, good.
**John:** No labor actions will have occurred. This would not be a covered project. I think we’re all good here.
**Craig:** The only reason I mention this is because I actually became a member of the AMPTP yesterday.
**John:** Wow, that’s really good. You’re now a signatory company. That’s great.
**Craig:** I’m a signatory company, yeah.
**John:** Craig, can you fill out the whole thing through? Did they hotbox you? I feel it’s like joining a fraternity, right?
**Craig:** You have to sign a document that says, “I hereby forswear my soul.” I am not a member of the AMPTP. It’s a fun idea to think about how you get jumped into that gang though.
**John:** It’s got to be fun.
**Craig:** Just men in suits punching feebly at you before you all get on your private jet to another billionaire’s event.
**John:** Absolutely. Tim Cook had to be jumped into it. I bet it was a wild thing, because Apple wasn’t a part of this, with the negotiations before this, and now they’re in. Netflix is in. It’s gotta be a lot. Where do they take you? What do they do? Are you doing shots? Is there a goat involved in something? I don’t know.
**Craig:** No, it’s not that cool.
**John:** It’s not that cool?
**Craig:** It’s not that cool, no. I gotta imagine it’s a fairly gray affair.
**John:** It really is. All the autonomy you thought you had, no longer. You thought you could control your own industry? No, no, you have to join this cabal. We will now make our deals together.
**Craig:** You worked your whole life to become the most important person at this massive multinational corporation, and now you have to join a group where your competitors have a say in who you get to hire and how much you pay them. What a great deal for them.
**John:** Good stuff.
**Craig:** While this is airing, theoretically the Writers Guild is back at work negotiating with the AMPTP. Good luck, John. I know you’re on that committee.
**John:** Thank you. Other bit of news, Weekend Read, which is the app we make for reading scripts on your iPhone. This week, we put out a version of it that runs on your Mac. If you have the iPhone version and you have a Macintosh, or one of the most recent Macintoshes, anything with a Silicon chip in there, it just now runs on your Mac too, which is handy, because you can just drag scripts from your desktop or whatever into Weekend Read there. It’s free. If you’re using it on your iPhone, you also have a Mac, just go to the app store, and you can install it on your Macintosh. It’s handy. All those scripts sync in the background. You can read stuff on both.
**Craig:** Fantastic.
**John:** We have some follow-up, I see here in the Workflowy. Can you talk us through, Drew, what we have?
**Drew Marquardt:** Sure. In Episode 605, we were talking about how racist characters are ultimately a little bit boring and aren’t a redeemable racist.
**John:** We talked about it’s very hard to find good examples of redeemable racists. It looks like we were wrong. One of our listeners, tell us.
**Drew:** We had a few people write in. Jafat wrote, “Although it is not his primary characteristic, Jack Nicholson’s character in As Good As It Gets is not only a racist, but an overall bigot throughout Act 1 and into Act 2, and yet he’s still a delightful protagonist to follow. The script builds so much empathy with his OCD condition, and we understood his bigoted tendencies to be a manifestation of his insecurities. So I guess great acting and great writing can make a racist appealing sometimes.”
**Craig:** I’m going to go ahead and reject that. I’m going to reject it.
**John:** Why are you rejecting it?
**Craig:** Love the movie. Jack Nicholson’s character is defined by his pure misanthropy. He hates everyone, everyone except a dog, who is not a person. He hates people who are a different color. He hates people who are a different gender. He hates people who are the same color and gender. He hates people are a different sexuality. He hates children. He hates everyone. The problem there is, if we’re looking for a redeemable racist, his racism is basically one wedge of a massive wheel of I hate everyone. He especially hates himself. I’m going to reject that.
**John:** Luckily, we have other examples here that you can choose to accept or reject.
**Craig:** Great.
**John:** What else do we got there?
**Drew:** We also have one from Laura P., who says, “You guys missed the most obvious example of a racist character who becomes redeemed: Green Book. Viggo Mortensen plays Tony Vallelonga, who grows and abandons his racist beliefs over the course of the movie. I know there’s a lot of hate out there for this movie, but I liked it. I saw it at TIFF before any of the hype, where it won the Audience Choice Award too.”
**Craig:** I’ve never seen it.
**John:** I think I saw half of Green Book.
**Craig:** You saw Green, or did you see Book?
**John:** I saw the Green part of Green Book. I remember the part that I did see. Viggo Mortensen’s character is driving a Black character through the South.
**Craig:** Mahershala.
**John:** Mahershala Ali. I don’t know it well enough. Yeah, there’s an arc there, but I don’t know. I’m not going to fight over a movie I don’t remember well.
**Craig:** Exactly. If you saw only half of it, you wouldn’t have gotten to the redemption part.
**John:** That’s true.
**Craig:** The movie you saw was a racist drives a Black man around, and then the movie ends.
**John:** The movie I remember seeing, the first thing that jumped in my head was not that Viggo Mortensen’s character was racist. I didn’t think of that as being his defining characteristic. Maybe that was because the whole movie was set in a place where racism was going to be so pervasive and dangerous that I saw him as being on the side of the Black character, even though they would grow and change over the course of the movie. I don’t remember it well.
**Craig:** We can’t push back on that, Laura P.
**John:** Last example I see here is for Scrooge. There’s David G writing in, “I gleaned it could be almost untenable to build a movie about a racist’s path to eventual repudiation of the racism. I’d be curious to your thoughts on why the tale of Scrooge works so well.” Scrooge hates everybody.
**Craig:** Scrooge has disconnected from people. His disconnection from people is, we ultimately learn, an extension of events that occurred in his youth. Everybody can identify with that. Everybody’s had moments where they felt alienated from the people around them, disconnected from the people around them. Everybody’s had moments where they were viewing other people through the prism of pessimism. People will oftentimes lose their naïve, childlike, wondrous spirit. It’s universal. It’s so universal that we simply cannot stop telling Dickens’s story to each other in 4,000 different ways.
**John:** Scrooge is not really that different than Jack Nicholson’s character in As Good As It Gets. Scrooge hates all people. He’s a wealthy version of Jack Nicholson’s character from As Good As It Gets.
**Craig:** Basically, yeah. They add a little mental illness flavor to Jack Nicholson, although I will say, were you to make As Good As It Gets today, I certainly would not be putting so much weight on his obsessive-compulsive disorder, because it just doesn’t work like that. Obsessive-compulsive disorder is not connected to misanthropy. I don’t even think the movie made a good argument for it to be connected. It just was also there.
**John:** More follow-up from the same episode. This is about Lonesome Dove.
**Drew:** Matthew writes, “I just wanted to follow up about the mention of Lonesome Dove first being a script which Larry McMurtry then turned into a novel, which then got turned into a mini-series. Craig said the idea that such a massive novel had started out as a 120-page script terrified him. I thought the same, so I checked, and it didn’t. It started out as a 288-page script. It originated in a film called The Streets of Laredo, which was intended as a vehicle for John Wayne, Henry Fonda, and James Stewart. The 288-page script was written by McMurtry with Peter Bogdanovich in 1972. The project failed to materialize, and McMurtry eventually chose to expand the idea into a sprawling 843-page novel. The length might have had something to do with the film not getting made in 1972, despite McMurtry already being well-established at the time, having written the movies Hud and The Last Picture Show. I imagine the success of his 1983 film, Terms of Endearment, was instrumental in getting the Lonesome Dove mini-series made.”
**Craig:** Wow. I’m not saying that they were coked up when they wrote a 288-page script in 1972. However, it wouldn’t be surprising if they were.
**John:** Let’s think that through that, because at some point, they had to realize, how long is this thing? This is 288 pages. That’s more than two full movies.
**Craig:** You’re the voice of anti-cocaine.
**John:** I know. I’m sorry. If any folks on the podcast-
**Craig:** By the way, I’ve never used cocaine.
**John:** I’ve never used cocaine in my life.
**Craig:** From what I understand, and I’m not being facetious, I really haven’t ever used it, but from what I understand, when you are on cocaine, you don’t stop and say, “Uh-oh, this might end up being 288 pages.” It’s more like, “Yeah! Yes, more. We’re actually reinventing cinema, man. Go.” On the other hand, 288 pages is a drop in the bucket compared to where it ended up, which was 800-and-some-odd pages. It may have been that this was trying to be a novel the whole time. The idea of sitting with somebody and hitting…
Even Scott Frank at his most lengthy first-draftness I don’t think is ever going to approach 288 pages. Scott is infamous for… It’s not even infamous. It’s just his process. His first drafts are always really, really long. I don’t think he’s ever hit 288.
**John:** That’s a lot. It also could’ve been a problem of just the time. It was an era before mini-series, probably. I’m trying to think. First mini-series I remember is Roots. That may not be the actual first real mini-series.
**Craig:** That was the first modern mini-series, or proper one, and that was ’76, ’77, ’78, something like that.
**John:** This is ’72, so it’s predating that. It does feel like a mini-series is the right way to tell a story sprawling, or a book. It found its way.
**Craig:** I’m glad it did. It’s a hell of a book.
**John:** Well done, everybody. Let’s talk about words. Craig, we’ll start with your loan out company I now know, because I see the end credits of your shows, is Word Games, correct?
**Craig:** Let’s be specific. It’s not my loan out company.
**John:** Sorry. It’s your production company?
**Craig:** It’s my production company, exactly, because I have a different thing. Loan out companies, I think we’ve talked about this on the podcast a number of times, are really just doing business as type things for the purposes of income and taxation and so forth. Word Games, yes, it’s my production company. I don’t mean to dress it up like I’m running Bad Robot or something. I’m aware I’m not. So far, Word Games has made two things, Chernobyl and The Last of Us. I’m a big fan of words and word games.
**John:** Obviously, what we do for a living is moving words around. On this podcast, we talk all the time about scripts and sequences and scenes and paragraphs and sentences, which are all built out of words. I don’t think we’ve done a segment just digging into what even are words, as an atomic unit that everything else is built out of. In Three Page Challenges, we may note that somebody’s using a word incorrectly, or just that there’s an odd choice of a word. We haven’t really dug into the words themselves. Obviously, probably half your recommendations are something related to puzzles, crossword puzzles, or other things you enjoy doing. Those are all built out of words as well.
**Craig:** There’s usage and vocabulary and definitions and things like that. Those are easy. Where the fun of words, the love of words comes, in terms of what we do, is very much connected to the intangible joy of the thought organization they imply and demand. Our thoughts are amorphous. Words solidify them. In fact, there’s quite a few theories of consciousness that argue that consciousness is a, I don’t want to say side effect, but words are a prerequisite for consciousness, that consciousness is formed by the mental manipulation of words.
**John:** Without language and words to organize thoughts, you don’t really have consciousness in the same way. We might talk about animals who we notice seem to be able to do certain things, they also have language abilities. There’s a reason for that.
**Craig:** Language is obviously separate from words, because there’s words themselves, but then there’s also grammar and the notion of inherent grammar, which is what Chomsky became famous for. The words themselves are yummily wonderful. I know I just said yummily.
**John:** Yummily. I understood what you meant, even though it was a word you-
**Craig:** Bingo.
**John:** It’s not the first time that word was ever mentioned or used in the world.
**Craig:** No question.
**John:** We can put the pieces together, in order to form a word like yummily and we get what it is. I see here in the Workflowy, meta crossword mechanism. Talk me through what that is and why it’s fascinating to you.
**Craig:** In thinking about this topic, because one of the things we’re going to discuss here is the evolution of computer language and these large language models and how computers are beginning to put words together in advanced ways, there are ways of putting words together that form the basis of games. A lot of word games really come down to playing word association. Have you ever played Decrypto, John? I can’t remember.
**John:** I love Decrypto.
**Craig:** Decrypto’s amazing. Without getting into a long description of the rules, what you’re trying to do is give your team a clue word that will help them identify which of four team words you’re pointing at. The other team can hear the clue word. You’re trying to basically not give away to the other team what your target word is with your clue word. It really is a game of sideways synonyms.
I was playing once with my friend Dave Shukan, who’s a word genius. One of our target words was tower. For one of the clues, I remember he said to us, and again, the other team can hear this, I want to say it was something like “flatbed.” We understood, after a little bit of thought, that flatbed pointed to tower, because what he was doing with that clue was saying that tower’s not tower, tower [TOU-uhr] is tower [TOH-uhr].
**John:** Oh, wow.
**Craig:** Same word, different pronunciation. That kind of strange association is delightful. When it occurs, there is a moment of joy. The example that I wanted to cite from, we’re recording this on Friday, August 4th, so the prior Friday was the fourth Friday of the month, which meant a difficult meta crossword from the great meta crossword master, Matt Gaffney. Of course, we all know Matt Gaffney.
**John:** Legendary.
**Craig:** He really is. He’s legendary. He’s a legendary constructor of meta crosswords. Those are crossword puzzles where after you fill in the grid, there is a hidden puzzle, and you have to figure out what the hidden puzzle is and what the answer is.
In this particular one, the gimmick that you eventually figured out was what he was doing was in certain clues, the beginning of the first word or the first couple of words in a phrase was either a single-letter or two-letter abbreviation you’d find on the periodic table. The trick of what he was doing was, if you expanded that out to the name, the full element name of what that symbol was, but then kept the rest of the word, it would become something new. For instance, the words “cut one,” if you take that Cu, which is the chemical symbol for copper, and you spell it out, “cut one” becomes “copper tone.”
**John:** Great.
**Craig:** I liked that one, but it didn’t necessarily delight me as much as this one. Let’s see if you guys can get this one. Now that you know the gimmick, it shouldn’t be too hard. “Fame singer Irene.”
**John:** Irene Cara is-
**Craig:** Okay, so the answer is Cara, now what do you with that, following this method?
**John:** Ca is calcium.
**Craig:** Good way to start. Calcium unfortunately isn’t a word. What else could you do? How about you try a single letter?
**John:** Oh, so carbon. Carbonara.
**Craig:** Carbonara.
**John:** Great.
**Craig:** Now, that-
**John:** That is delightful.
**Craig:** Damn. This is the sort of thing I occupy myself with all the time, which I love. I appreciate people like Matt, who can think of these things and then execute them so beautifully. Relevant to our discussion today, there is something gorgeous about the creation aspect of that, that words are something that we use to fill our minds and our speech and communicate. Also, there is a creative aspect to the manipulation of them that requires pleasure.
What we do as writers is smashing words together to create context and information and communication, but specifically also pleasure, enjoyment, even if the pleasure and enjoyment is the pleasure and enjoyment of crying. That goal requires another dimension beyond just the pure computational understanding of how to put words together.
**John:** It requires attention. It requires the ability to have a desire for what you’re going to try to do and the ability to anticipate how it will manifest in the brain or the mouth of the person who’s going to be experiencing it. That requires attention. You don’t get there accidentally. You have to be thinking about what series of words is going to create the effects that I want to get and why am I trying to create that effect. There’s a deliberateness to it. That’s what writing is. It’s the deliberate effect you’re trying to create by putting the words together in this order.
**Craig:** Precisely. I don’t want to be the “AI’s not going to take over,” guy, because everybody’s very invested in the thought that AI’s going to eliminate us all. However, if there is one thing that is insulating us from being eliminated by AI and the increasingly complicated versions of ChatGPT and other large languages, it’s this. It’s that large language models require prompts. We do not. We create our own prompts. Our prompts are prompted in turn by our wants and needs. Wants and need are prompted in turn by the pleasure principle. This is an interesting area.
We know, for instance, that calculators are infinitely better at operations than humans. Any mathematical operation you can come up with. The simple ones are addition, subtraction, multiplication, division, but there are more advanced ones. Whatever they are, it doesn’t matter what they are, a simple $5 calculator is infinitely better at it than humans. It can calculate those things faster than you can blink. Much faster.
However, humans appear to be infinitely better at proving conjectures than computers. Proofs for complicated theorems, theorems that have been sitting out there for centuries, and then eventually get proven. They don’t get proven by computers. So far, I don’t believe any computer has ever proven anything complicated. They can prove things that we teach them to prove, but that means we proved them first.
**John:** I will look up and try to put the link in the show notes to… I can think of, there was one mathematical thing that was scrawled in a margin, which they actually brute forced it with computers and basically eliminated all the other possibilities, therefore were able to prove it. Then again, that required human intention and intervention to figure out what are the things we needed to knock out in order to get to this proof.
**Craig:** Certainly, we can use it as a tool, but what we can’t do is say, “Here’s Fermat’s Last Theorem. Computer, prove it.” No. No. It took a human to do that. One method follows a process, and the other method invents a process. I do think what we do is about inventing new uses for words and concepts, and the collision and transformation of those words and concepts. That said…
**John:** That said…
**Craig:** Wow, have they gotten, they meaning the computers, gotten good at following processes.
**John:** The article we’re going to link to is by Tim Lee and Sean Trott. It came out this last week. It’s a really good detailed, but not actually too super geeky look at how large language models actually work. I think one of the important takeaways is that we understand the process of building them.
We cannot, in any given example, tell you how they got to the answer they got to. That black box is not by design. It’s just the nature of process, of how it’s stringing those words together. We can’t tell you how it got from A to B to C to D. It’s very, very difficult to show how that happened.
Let’s wind back and actually talk through some of the workings of these large language models, because it’s much more subtle and complex than I thought. We have these arguments like, these are all plagiarism machines. Plagiarism is passing off other people’s work as your own. While it’s true that all these models are trained off of the internet, so they are hoovering in all the stuff that’s been out there, the ways they’re stitching these words together is so different than even where I thought. I always knew it was all based on probability of what the next word would be, but how it gets to those probabilities is actually really fascinating.
I think it’s partly because our understanding of words is very different than the computer’s understanding of words. I wonder if maybe our own understanding of our own consciousness will eventually be revealed to be something a little bit more like what these large language models are doing, because it’s fascinating. They don’t treat words the same way that we would expect. They’re not thinking about definitions or spelling or its origin. It’s just these mathematical vectors, these points in this thousands-of-dimensions word space that is hard for us to even imagine.
**Craig:** I think that probably this is how we do it, neurologically. Even though the complexity, even though it seems like it’s pretty good, it’s nowhere near what we can do, because our neurons number in the billions, I believe, and they’re not quite there yet, but I guess eventually they will be.
It’s very relational. Words are defined as items that relate to each other, so a little bit like the way I think of characters. Character is meaningless without relationship. It’s just nothing. Similarly, words are meaningless without relationship.
The way the LLMs seem to work is by defining relationships and relationship strength in multiple ways, between a word and every other word. Some of those relationships are defined by similarity of definition. Some of them are defined by being opposite. Some of them are defined as, if this word is next to this one, then this one is likely to have this meaning, as opposed to this other alternate meaning. Everything gets ranked.
Look. Ultimately, neurons are on/off switches. It’s a little more complicated than that. There’s levels of ons and levels of offs, but you’re really dealing ultimately with a lot of circuitry there. I can see a world, and maybe it’s soon, where we can create an LLM that has the same capacity that we do, purely for language.
However, that’s not all we have. What we’re building with these things is, we are approaching the neocortex. That’s what we’re working on here. There’s a couple areas in the brain, like Wernicke’s area and Broca’s area, that do a lot of this work. We’re approaching replicating that somehow.
What we don’t have, I don’t think we do, and I hope we never do, is what we have in our brain. I don’t want to ever give this to computers. That is a combination, a feedback system between a neocortex and a paleocortex, between the old brain, between the limbic system and the neocortex. This is where all of our danger is, but it’s also where all of our wonder is.
This is the most hacked, tropey thing that aliens will say after observing us for a while. “Strange, you know. Humans, you know. The source of your greatest flaw is also the source of your greatest value.” Yeah, basically. Which is that we are still animals, but we are enlightened animals, and we therefore can create works of art. The only reason to create a work of art is to hack the neocortex, to appeal to the limbic system. I don’t know why a computer would even bother to do it. It doesn’t have a limbic system.
**John:** It again doesn’t have pleasure.
**Craig:** Correct. It doesn’t have pleasure. It doesn’t have fear. Although it was interesting, I watched 2001: A Space Odyssey at the Hollywood Bowl last night, LA Philharmonic doing a gorgeous job, the percussionist living his best life, dong, dong, dong, dong.
**John:** Oh, of course.
**Craig:** When the HAL 9000 is being turned off, slowly, memory bank by memory bank is how they did it, he repeated the phrase, “I am afraid,” which was really interesting. You could say to a computer right now, “Okay, ChatGPT, write about having to go on a first date, but make sure that you explain how afraid you are and why.” It can do that, but it won’t be afraid. That I think is maybe what’s going to insulate what we do from what that does.
**John:** Let’s circle back to the words of it all, because as humans, as we deal with words, if we want to organize words, we would put them in a dictionary and alphabetize them, figure out their definitions. It’s not how these large language models are working.
Instead, what they’re doing is… You can think about it like a globe or a map. Just in the same way that two cities might have longitude and latitude, every word is going to have a longitude and latitude for where it fits in this space.
Google did this in 2013, this thing called word2vec. What they did is they scanned zillions of documents to figure out what are all the words and what are those words near, based on what those words are near, how can you find some meaning, the probability that these two words will occur together.
When you do this, you can do some kinds of math that are really interesting. If I say Berlin minus Germany plus France, you can get to Paris. Take this one thing away from Berlin, and then add France, oh, the relationship must be that Paris is related to France in that way. We get that. It’s like the analogies we had to do in the old SAT. It could learn… Learn sounds a little too intensive, but patterns will emerge. A cat is very near a dog, and so therefore, if something is true about a dog, like a dog goes to a vet, then a cat will probably go to a vet. That makes sense.
So much of what’s fun about the word games that Craig likes to play are double meanings of words or homonyms and other things. Tower was an example, tower [TOH-uhr] versus tower [TOU-uhr]. Subtle things, like this article points out, like Mary works for a magazine, Bob reads a magazine. They both have the word magazine, but those are really not quite the same word. The first one is a company that she works for. The second one is a physical object. Those two separate words would be close to each other, but they’re not the same thing. It’s a part of speech. They really are different. They’re different things and may be represented as different vectors, so a huge grid of numbers that indicate where in this word space you put those two different words.
**Craig:** Polysemy.
**John:** Polysemy. I didn’t know how to pronounce this. Thank you for saying that.
**Craig:** My pleasure.
**John:** When there’s many possible meanings for a certain word of phrase or combination of letters. What would surprise you about this article is how it figures out context. A sentence like, “The customer asked the mechanic to fix his car,” how do we know that the his is the customer rather than the mechanic? That’s fascinating, because it has to figure out what are the probabilities that that his, it knows that it’s a male pronoun, but how likely is it that it relates to the customer versus the mechanic? It’s just math and probability that gets us there.
**Craig:** Even if the connections weren’t formed the same way that these connections were formed, this, I suspect, is not far off from the way the connections exist in our minds, where things both inhibit and reinforce certain possibilities.
What we know, because we may laugh at ChatGPT for some funny mistakes, but we make mistakes all the time. People are constantly getting confused. People have entire conversations where somebody’s like, “Wait a second. Sorry. Are you talking about him or him? Because I thought you were saying him, he, but I think you might’ve meant him, he.” “Oh, yes, I was… “ There you go.
This happens all the time when we’re writing the… I was just talking to somebody the other day about this problem. Just within sentences, when you have conversations between two people who use she and hers pronouns, it can be a nightmare. This is where I’m not sure how you teach a computer this, because I think it’s the pleasure principle. For some pleasurable reason, we do not like when names are repeated over and over and pronouns are not used. Actually, it makes no sense.
**John:** It would be much more efficient if I referred to Craig all the time.
**Craig:** Correct. There’s really no reason to use a variable for a specific if you’re a computer, just none, unless in cases where a variable isn’t required. Why? Why don’t we like hearing our name over and over and over and over? We don’t know. We don’t know, but it just doesn’t feel good.
All that stuff is really fascinating to me. I do think that it’s not fair to say that these things are plagiarism machines any more than it’s fair to say that human beings are plagiarism machines, because that’s how we learn, by scraping all the language that’s pouring into our ears as our brains form in youth.
**John:** It’s true. That’s why when you grow up with a native language, and that is the language that you’re most comfortable using, and then when you use another language, sometimes you can use it natively, but it’s challenging to. Even the best translators, they will translate into their own language, but they won’t generally translate out to their second or their third language. There’s a reason why your native language is your native language.
I was recently in Africa. We were being driven from one place to another. There was this negotiation that happened between our driver and this checkpoint person. As we drove away, I was like, “Oh, can I ask you, what language were you using?” It was like, “Oh, I had to use three different things, because we were trying to figure out which one we both spoke.” That kind of code switching we see in Los Angeles all the time with Spanish. It was fascinating to see it in another country, where there weren’t just two languages, but 11 things that could be sorted through.
**Craig:** That’s why I think we aren’t too far off from, there’s already been demos of this, the Babel fish style earbud that hears a language, automatically translates it in real time, and then pipes it into your ear in your language. The fact that you can understand any language doesn’t mean you can speak it. That again, it’s just fascinating to me. It’d also be really interesting to see how those models handle subtleties.
There’s going to be a lot of misunderstandings and miscommunications in the early days of the Babel fish earpiece. There are going to be some people that stand up from a table and throw a drink in someone’s face, only later to find out, “Oh, sorry, my thing just mistranslated.” I’m there for that. I’m there for that.
**John:** It’s going to be exciting when it happens. Just to bring a D and D reference, in the D and D universe, there’s Common, which is the language that most people will speak, just for easy use. They also speak Elvish or Dwarvish or whatever else, but Common is the language.
I would say that on this last trip, you recognize, oh yeah, English is sort of Common in most places. When people can’t find another language to speak, they’ll go to English, because that’s the second language of so much of the world that they can get by. Even two people who grew up in South Africa might end up just defaulting to English because that’s the language that they know they can both get by in.
**Craig:** That’s right. That was not always the case. The lingua franca was called lingua franca because it was French. French was considered common back in the day. In fact, to reference 2001 again, there was a conversation happening between some people on board the space station that were from different countries, and they were speaking French, because that was the thing they can all agree on.
Over time, that has been absolutely replaced by English, not even by a little, but by a lot. There are all sorts of reasons for that, of course, economic and political. One of the things that English has going for it is that it is relatively easy, compared to other languages. There aren’t gendered nouns. Verb conjugation is very simple.
**John:** We got rid of almost all of it.
**Craig:** Pronunciation can be absolutely confounding, befuddling, but then again, there are a lot of people that are native English speakers who mispronounce words all the time. Look, polysemy, a lot of different ways of approaching that one.
**John:** Absolutely.
**Craig:** That’s a Greek word. Pronunciation can befuddle people. Have you seen that really funny thing they did, every ‘90s pop song ever?
**John:** It was great. Every European pop song?
**Craig:** Yes, every European pop song. One of the things that they’re really smart about is they’re implying that these singers were given lyrics in English, but because they don’t speak English and they don’t know how to pronounce things, they failed to see how certain things were intended to rhyme. He goes, “Boom, hear the bass go zoom.” That’s really funny. Boom and zoom, they do rhyme. That’s okay. People really have just settled on English. Let us count ourselves lucky in that regard.
**John:** Oh yeah, it’s imperialism. It’s all the things that happened. It’s the accidents of history that English became the [indiscernible 00:37:09] language, but there were worse choices out there. It’s a relatively easy language to speak. It’s messy to spell. There’s lots of things you could’ve improved. If you could get another crack at it, you could do a better job. If you could do it Esperanto, if everyone could speak Esperanto, it would be easier for everybody, but it’s not what’s going to happen.
**Craig:** What a silly dream of the ‘70s.
**John:** I love Esperanto.
**Craig:** Or ‘60s, I guess.
**John:** Craig, my other thing, as we wrap up this segment on words, one of the things I love about listening to podcasts is you have a lot of really smart people who are talking, and you realize that they are saying words out loud that they mostly would otherwise write. I just typed a word here. I’m curious how you pronounce this word that I just added to the Workflowy.
**Craig:** Let’s see.
**John:** Underneath the list of official languages of Africa.
**Craig:** Say this word. Subsequent [SUHB-suh-kwuhnt].
**John:** I would always say subsequent [SUHB-suh-kwuhnt], and then I heard someone on a podcast say subsequent [suhb-SEE-kwint]. I’m like, what? I took a note of it, and I looked it up, and that is an accepted pronunciation.
**Craig:** It’s accepted. It wouldn’t be the first listed pronunciation, but it’s not incorrect. If you said suhb-sih-KYEW-int, that would be wrong.
**John:** That would be wrong.
**Craig:** That would just be wrong. Or if you said SOOB-see-kwint. SOOB-see-kwint is terrible. Don’t say that.
**John:** Don’t say that. Any further takeaways from our word segment, Craig?
**Craig:** No, other than I suspect we are going to get quite a bit of listener mail from people that work in AI or with LLMs correcting us and scolding us for all sorts of things, to which I say, bah.
**John:** Bah. I say direct your criticisms to the people who wrote this really good article, Tim Lee and Sean Trott, because we were sort of summarizing what we read there. If we summarized it wrong, meh, sorry.
**Craig:** Bah.
**John:** We will put a link in the show notes to this article, which was really well done and did talk me through a lot of stuff.
**Craig:** It was a good one.
**John:** Did talk through a lot of stuff I’d never seen before.
**Craig:** It was really smart.
**John:** Listener questions. What do you got for us, Drew?
**Drew:** Ken in Norway writes, “Longtime listener here who finally transitioned to TV writing a couple years, helped and inspired by your guidance, so thank you for that. Now that I have to write stuff people actually are planning on producing, I keep thinking about Craig’s tip to not move until you see it. The way I interpret it is to not write the story or the scene until you believe it. My question is, how do you balance this with deadlines? You can’t really plan on having an epiphany Wednesday morning or finding the missing piece of the puzzle by the end of the week, so the only reliable way forward is to muscle through. On the other hand, you don’t want to churn out something that you feel in your gut isn’t really working. That won’t help anybody. Do you have any thoughts of how to better your odds of seeing it in time, or is the seeing it relative to the amount of pressure you’re under?”
**Craig:** This is one of the reasons it’s important to start working right away, because I think writers sometimes presume they have more time than they do, because they are not pricing in the moments where they can’t see it. If you get eight weeks to write something, start writing it on day one. Start doing the work on day one at least, if you’re outlining or whatever it is. Get to work.
When you know what you’re doing, well make hay while the sun is shining, because you’re going to run into some trouble at some point. You’re going to need the extra time to think things through. Hopefully, taking advantage of the time you have and pushing through and being smart and banking some time is going to help you with that.
The other thing is, there are times where you may need a little bit of extra time. Now, certain circumstances, you will not be afforded the extra time. In other circumstances, people will afford you some extra time. What they’re not telling you is that there are terrible costs to making changes. It’s going to take them a lot more time to find a new writer than it will for you to just finish, generally speaking.
What I would suggest is, from a practical point of view, if you’re going to call and ask for more time, when you do, explain that you had a problem, it took you some time, but you have solved it. You’re very pleased, and now you’re full speed ahead once again. However, this has put you a little bit behind. If you call and say, “I am having a problem. I cannot see it. I need extra time,” that’s just going to send everybody into a tizzy.
**John:** All that advice is fair and good and true. I will say there have been times in my life where I’ve had to just muscle things, where these pieces don’t quite fit the way I know they possibly could fit, I can envision a world in which these things worked better and I got from this moment to this moment a little bit more smoothly. The needs of getting this draft in outweighed the artistic pinnacle optimization of this one moment. Therefore, I muscled it and knew that we’re going to go back there and work through that again. That’s just the reality.
Then sometimes, the train is leaving the station, and you have to make that scene work the best it can work, so that you can actually keep going. Don’t beat yourself up about that.
Craig’s basic advice, basically, is to make sure that you’re using all of your time all the way up through it, is true. You have to do that. The only thing that makes me nervous about don’t move until you see it is people can, if there’s not a deadline, just be paralyzed forever. They can be trying to solve a problem that essentially has no answer, or they’re imagining a solution that doesn’t actually even exist, and we actually have to reevaluate what is causing the problem, rather than trying to find a solution to that problem.
**Craig:** I would say that’s part of not moving until you see it, because if it’s been a week and you haven’t seen it, you may not have the problem you think you have. You may have a very different problem. At that point, you need to step back even a little bit further.
You’re trying ultimately to follow the “a stitch in time saves nine” plan. By the way, when I was a kid, I did not understand what that meant, because there were no commas in it. A stitch in time saves nine. I’m like, what is a stitch in time?
**John:** A stitch, comma, in time, saves nine.
**Craig:** Where was the appositive phrase markers for me? They were nowhere. “A stitch in times saves nine” makes no damn sense. “A stitch, in time, saves nine,” makes… Commas. Commas, folks.
**John:** They’re so useful.
**Craig:** Damn.
**John:** So useful.
**Craig:** Hopefully we helped there, Tim.
**John:** Let’s try one more.
**Drew:** State Your Name and Outlet writes, “I was listening to the episode That’s a Good Question, and it made me want to ask you something I’ve been wondering for a while. Over the last few years, I’ve taken on more and more work as an interviewer for a film and TV publication, which has led me to participating in something I’ve heard Craig and many others talk about with contempt. That is, of course, the press junket.”
**Craig:** Dun dun duh.
**Drew:** “I’ve been a writer for most of my life, and have much more sympathy for people like you two, who are stuck in a room doing four-minute interview after four-minute interview, than I do for people trying to get a snappy quote for a headline. I try to ask interesting and crafty questions, but also don’t want to presume whoever is on the other end wants anything more than to get out of there. I guess my question is, what is your ideal junket interaction?”
**John:** That is a good question. Maybe you’re a good junketeer yourself, because you’re asking a good question here. I would say my experience at press junkets, the person who gets the assignment basically understands, “I understand what this show is, what this movie is,” and is genuinely curious and fascinated by some element of it, some specific thing. That is what’s going to lead to a good answer from me, because I’ve just been saying the exact same thing again and again and again, but if you can come at a specific and ask me those specifics, it will go well.
I think back to when Greta Gerwig came on the show to talk about Little Women. It was the first time we actually had script pages in front of us. I could ask her, “Talk to me about this scene and why you did it this way. Was there any push back on that?” No one asked her specifics about a scene or literally the words on a page.
If you can come to one of these things and ask a specific question about a specific moment, you’re going to get an answer that is going to be so much more useful to you and will actually delight me or Craig or whoever you’re sitting across from.
**Craig:** Listen. I don’t want to imply that I’m contemptuous of the people that do these things. I’m not. They’re doing a job. It’s the overall experience that’s exhausting. Each individual person is doing their job.
The first person of the day is the best person, because you’re fresh. You started. They’re asking you a question. No one’s asked you that today. You give your answer. Sometimes the best person of the day is the last one, because you know you’re about to be done, and also, by that point, you’ve answered that one same question so many times, you’ve found the perfect mix of words to answer it.
The process can be exhausting and dispiriting for me, because you are being often asked three questions over and over and over. My advice for State Your Name and Outlet is to really think carefully if you need the answer to that obvious question, because after all, while everyone else is asking it, it means everyone else is going to have that content in their article. What if you just refuse to ask those questions, because no one needs the answers to those questions.
Honestly, nobody interviewing me today needs me to answer the following question: What was the hardest challenge to adapt a video game into a television show? I’ve answered this question 4 million times. If you are curious, you can merely Google, and you’ll have the answer in many, many slightly varying versions.
A new question is a delight. You will see the person on the other side of the microphone light up if you are asking them a novel question, and particularly, if you are asking them a novel question and studiously avoiding asking them the one that they’re asked over and over and over. That’s my advice. Just don’t ask those questions. You know what they are. Everybody knows what they are. If someone above you is saying, “You gotta get them to answer this question,” push back a little bit and say, “Do we?”
**John:** You don’t have to ask that question, because honestly, you’re probably not going to print the question. You’re going to print what Craig says. If you get Craig to say something interesting, that’s what you’re going to print. Then you can frame it however you want to frame the story, however you want to frame it. Ask something interesting, and you’ll get an interesting answer.
**Craig:** That’d be an interesting exercise, to show up at a press junket with your prediction of those questions written down on cards, and just put them out in front of you, but face down. If someone asks that question, you just reach over, because they’re numbered, and you know which one they are, and you just turn it over, and that’s how they know, “Uh-oh, I should probably move on to another question.” Boy, would you be a jerk.
**John:** Oh my god.
**Craig:** Oh my god.
**John:** That asshole.
**Craig:** Literally every interview, every article would be about how you’re just a dick, because you know what? You would be a dick. Listen.
**John:** It’s the game. You’re not playing the game.
**Craig:** When somebody said, “What was the trickiest challenge to adapting a video game to a television show,” never once was I like, “Oh god, really? I already answered that question.” No, I answer it like it’s the first time I’ve heard it, because that’s the polite and nice thing to do.
The whole purpose of doing that stuff is to give these folks something to run in their publication which theoretically will appeal to people at home to turn the show on and watch it. It’s a commensalist relationship. You can’t be a jerk.
By the way, I have no problem with people who get up and walk out of interviews where the interviewer is being a jerk. That’s different. I’ve seen some of those. Happily, I’ve never had that experience. There’s no need to be a jerk about any of this. That said, since State Your Name and Outlet asked for the advice, the advice, don’t ask that question.
**John:** It’s come time for our One Cool Things.
**Craig:** Yay.
**John:** My One Cool Thing is a book that I read over this last couple weeks. This Is How You Lose The Time War by Amar El-Mohtar and Max Gladstone. I love a book that’s written by two people, which is unusual. It is this sci-fi romance between these two agents battling on opposite sides across these multiple timelines. It’s structured as an epistolary, so these letters being exchanged back and forth between these two strangers, who then get to know each other over the course of this.
It is sprawling. The world-building is amazing. The word usage, it’s good for this episode. Just the word craft is incredibly good. The voice and the difference between the voices of these two agents is great. I just really dug it. It’s one of those high-concept things that you think couldn’t work, and worked brilliantly.
The backstory on it is also fascinating, because this came out in 2019 and got some acclaim, but I only heard about it because this last May, this guy named Bigolas Dickolas Wolfwood tweeted about this, basically said, “Stop what you’re doing and pick up this book. It’s amazing.”
**Craig:** What a world we live in.
**John:** That went viral. That’s how I got aware of it. I picked it up, and I’m really glad I did. This Is How You Lose The Time War.
**Craig:** My One Cool Thing will flow right into our Bonus episode, as promised. In memory of Sinead O’Connor, who passed away, unfortunately, far too young, just a little over a week ago, I believe, I want to recommend one song to watch, because I think the performance is just as important as the song. Visually, to see it is just as important.
It’s a performance of her song Troy. I don’t know if I would call it a deep cut, but it wasn’t a hit. One of my favorite songs. It’s a live recording from a concert in 1988. It’s a long song. It’s worth the time, because what you see on display there is not only beautiful songwriting and excellent grasp of words and lyrics, but honesty and emotion and performance.
When we talk about, “Oh, I don’t believe that,” or what is quality, the genuine expression of emotion, and the emotion that is expressed changing over the course of the song, from begging to praise to condemnation to confusion to self-doubt and recrimination and finally to accusation is remarkable. It is kind of like what we do when we’re doing our work at our best. It is astonishing. When you get to the end of the song, I challenge anybody to neatly summarize her relationship with the person she’s singing the song to. You cannot. That is why it’s great.
**John:** It is great. A thing that I really admire about that song is it doesn’t have a chorus verse structure at all. It keeps building and building and building. You don’t know where you’re at in the song. Then it gets up to this big finale. It’s fantastic.
**Craig:** It’s stunning to watch. There’s a couple moments where the song will take a break, and then it begins, like a pause, a pregnant pause, and then it starts up. In the first one, as is often the case in concerts, you hit this big note and a pause, and the pause is filled by people in the audience going, “Woo!” The second time through, the other ones, no one says a goddamn thing, because they’re transfixed, as well they should be.
**John:** As they should be. That is our show for this week. Scriptnotes is produced by Drew Marquardt.
**Craig:** What what.
**John:** It’s edited by Matthew Chilelli.
**Craig:** Whoop whoop.
**John:** Our outro this week is by Jacob Weisblat. If you have an outro, you can send us a link to ask@johnaugust.com. That’s also the place where you can send questions and follow-up. You can find the show notes for this episode and all episodes at johnaugust.com. That’s also where you’ll find the transcripts and sign up for our weeklyish newsletter called Inneresting, which has lots of links to things about writing.
We have T-shirts and hoodies, and they’re great. You’ll find them at Cotton Bureau. Looks like we have a new T-shirt that’s going to be coming up soon. Craig, you’ll love it. I’ll talk to you about it off mic.
You can sign up to become a Premium member at scriptnotes.net, where you can get all the back-episodes and Bonus Segments. If you are listening to this on Tuesday, you can still join us for the live show on Wednesday. You can get a streaming ticket for that. You can see us and our special guests at The Dynasty Typewriter. Craig, Drew, thanks for a fun show.
**Craig:** Thank you, John.
[Bonus Segment]
**John:** Sinead O’Connor. I feel like maybe we should do a little bit of table-setting here, because I recognize that we are both men in our 50s, and she broke on the scene when we were transitioning from high school to college. We were just exactly the sweet spot for Sinead O’Connor to be a thing. My daughter has no idea who Sinead O’Connor is or was. Craig, what can you set up for us about her?
**Craig:** Sinead O’Connor released her debut album, The Lion and the Cobra, it was 1987, so I was a junior in high school. I suspect you were-
**John:** As was I.
**Craig:** Oh, you were a junior as well. Then followed that up with I Do Not Want What I Haven’t Got in 1990. I’ll readily admit that I obsessively listened to those albums and dropped off after that. Those were the albums that I listened to.
**John:** Those were the albums for me too.
**Craig:** I listened to them all. My friend Gene and I had rooms side by side in our dorm. We would play Gin Rummy, because that was more fun than doing work. We would just either have The Lion and the Cobra or I Do Not Want What I Haven’t Got. What’s interesting is you’re like, oh, here’s a couple of bros in college listening to a woman who was unlike anyone else at the time.
Maybe you could argue that Boy George had kind of set a little bit of a stage, as well as Annie Lennox, for somewhat androgynous pop stars. Sinead O’Connor, it didn’t feel like a gimmick necessarily. It wasn’t structured. She was an Irish woman who did come out of nowhere and exploded out there. She was notable for shaving her head. She had no hair. She was also stunningly beautiful. It was this really weird, confusing combination of elements. She had no problem both using and denying her own physical beauty to center and put forward her songwriting ability, her music, and her voice.
Her voice, this is where, hey kids, go ahead and listen to Troy live and see what people used to sound like when there wasn’t auto-tune or backing tracks. Her voice was astonishing. She could do things with it, and not in a Freddie Mercury virtuosic way, but in an emotional, expressive way, that I haven’t actually heard anybody else do.
**John:** She had a dynamic range that was incredible too. Her ability to whisper a lyric and then just belt it to the far realms was incredible. She did feel unearthly, and partly because of the shaved head and all that stuff. It did feel like she was some elfin creature who’d come out of some other dimension. That also carried through into her voice.
There’s a tradition of incredibly talented female singer-songwriters, Joni Mitchell and going way back. That’s not new. I think about her as a template for other people down the road. You can think of Lorde or Billie Eilish. You don’t want to diminish what makes each of them unique and special.
When I think of her in addition to those, I worry for women who would build up to these paragons now, that the same kinds of things could happen to them that happened to her, because her arc is not great. She had this success, and then she had struggles with mental health. She had lots of things that later in her life became real challenges.
Obviously, the story everyone knows about is that when she tore up the photo of the Pope on Saturday Night Live, which I remember live and not really understanding what even happened. I didn’t even see it was the Pope. I didn’t understand what that was in the moment. That became a source of huge controversy.
**Craig:** It is undeniable that Sinead O’Connor struggled with mental health. That said, so many people do. That wasn’t why she got knocked off her perch so hard. It was that moment.
Sinead O’Connor was placed with, I believe it was a convent-run laundry, in Ireland. These were notorious Catholic institutions that were abusive. The Catholic church has a dreadful track record, as we all know, when it comes to care-taking of children and doing damage. She had a tremendous resentment against it.
If today, someone who had been raised in a Catholic orphanage went on television and ripped up a picture of the Pope, I think there would be a healthy and earnest debate about it. What there wouldn’t be is a consensus that she was a nutter, because that’s what happened. The very next week, as I recall, on Saturday Night Live, they made fun of her, which was easy to do. Put a bald wig on a lady and everyone immediately starts laughing, like, oh, Sinead O’Connor, she’s ripping up a picture of the Pope.
The thing about Sinead O’Connor in retrospect was, A, she was right, and B, she didn’t seem to give a shit. Even though it was within the framework of having the number one album in the United States, her second album, I Do Not Want What I Haven’t Got, went number one, and with this massive hit song written by Prince, Nothing Compares 2U, she was also still the woman who was shaving her head. She didn’t care. She took a big shot there.
I don’t think ultimately that that incident is why she suffered mental health problems. I do think that incident impacted her career. To the extent that that hurt her and upset her, that is deeply unfortunate, because it shouldn’t have. Catholic church is a mess. I don’t care. People can go on Saturday Night Live and make fun of me. I don’t care. I also don’t care, because I’m so brave and also bald. Easy for me to say now. Not easy for anyone to say back then, but certainly not a young woman on TV, who was already turning away from convention.
**John:** You already mentioned Troy as a song people should seek out. I’m going to put Success Has Made A Failure of Our Home, so Loretta Lynn’s song that she covered on her third album.
**Craig:** Am I Not Your Girl.
**John:** Am I Not Your Girl. It’s fantastic. It’s a song I was not aware of until I heard her sing it. It’s gorgeous. One of the things I loved about, there was a moment in time where there would be, Red Hot and Blue, there were a bunch of special albums, where they would bring in pop stars to do one charity album. You’d hear these great bands do a cover of something that was completely not in their wheelhouse. Those were fantastic. There’s a Cole Porter one. It was Red Hot and Blue. I just remember that CD being just a delight, top to bottom. This was again this case where to hear her cover this song was fantastic to me.
**Craig:** I’m not sure there would be any song that if someone said, “Do you think Sinead O’Connor could successfully cover this?” I think the answer is always yes, again because of the elasticity of her voice and the rawness of the emotion and the honesty. Just to refer once more to that performance of Troy, it’s about commitment. One of the things that we look for when we are guiding actors as directors is commitment and an abandonment of self into the moment. It would’ve been really interesting to see Sinead O’Connor trying that form of art, because her commitment while singing is 100%. It is pure. It is not manipulative. It’s just true. It’s just true. Really interesting.
Also, the other thing that happened with her later in life, which led to I guess some more snickering or something, is that she converted to Islam and stuck with it too. It didn’t seem like it was a whimsical, “I’m nuts, let me try this thing now.” She stood with it. We’re using her name, Sinead O’Connor. That’s the name she was given, but that is not the name that she ultimately chose to live as. She was born Sinead Marie Bernadette O’Connor. Side note, nobody until Sinead O’Connor understood how to pronounce any Gaelic word at all.
**John:** It’s true.
**Craig:** When it first came out, everyone was like, “Who’s sin-EE-ad?” You’re like, “No, man, it’s shuh-NAYD.” “What?” That’s how we learned how that worked.
**John:** Like the classic spelling of Sivan, no one was able to do that before.
**Craig:** Nobody was able to do that before. She did convert to Islam and changed her name to Shuhada’ Sadaqat, important to acknowledge that.
**John:** Absolutely. The other thing I want to throw, I fell down a YouTube hole with Sinead O’Connor, just looking at different performances, a more recent one is All Apologies, the Nirvana song. She does, of course, an amazing cover of that.
**Craig:** That’s the thing. I feel like, oh-
**John:** There’s a kinship between Kurt Cobain and Sinead O’Connor that makes sense.
**Craig:** I think there’s a kinship between every artist that maybe suffers from an overcommitment. We in the audience are the beneficiaries of those who commit so fully that they burn themselves up in the flame of their own feelings and art. It’s painful to watch how many people have done that. In our own ways, as writers, because what we do is much slower and not performative, we are insulated from that, but not completely. I’m sure you’ve felt it. I feel it all the time, the price. There’s a price.
**John:** Craig, one thing is we have to balance though. We can recognize that there are performers who do that and that that is a real thing that happens, and being careful to not over-glamorize it to the point where that’s the only way to make art is to destroy yourself to it.
**Craig:** It’s not. It’s not. That’s the thing. What happens is, I think it is exciting when you see people throwing themselves into things so fully. By all accounts, watching Jim Morrison with The Doors in the ‘60s was mind-blowing. I completely agree with you. It is not necessary to make great art whatsoever. Disciplined artists ultimately win out as superior, because they are able to create more for longer, but to the extent that Sinead did do something unique. I don’t believe there has been anyone else like her since she burst onto the scene.
Also, one of the things that her passing drove home for me is how time screws with you as you get older, because when we were in college, Sinead O’Connor was so much older than we were. She was five years older than we were.
**John:** That’s all it was.
**Craig:** Now it’s like, she died, she was 56. I’m 52, but basically I’m 56. What’s the difference? It is remarkable how close we were in age. It makes it more upsetting. It makes it more upsetting.
**John:** It does. Rest in peace to Sinead O’Connor. If you’ve not checked out her music, if you’re a younger person who just barely knows the name, worth spending some time digging through that, because you will find an amazing artist and really just a ground breaker.
**Craig:** No question. She was one of a kind and gone too soon and will be missed.
**John:** Great. Thanks so much.
**Craig:** Thank you.
Links:
* [Scriptnotes LIVE! at Dynasty Typewriter in Los Angeles](https://www.eventbrite.com/e/scriptnotes-live-tickets-674019238687) benefitting [HollywoodHEART](https://www.hollywoodheart.org/)
* [Weekend Read 2](https://quoteunquoteapps.com/weekendread/), now available on [MacOS](https://apps.apple.com/us/app/weekend-read-2/id1534798355).
* [Large language models, explained with a minimum of math and jargon](https://www.understandingai.org/p/large-language-models-explained-with) by Timothy B Lee and Sean Trott
* [Word Vectors](http://vectors.nlpl.eu/explore/embeddings/en/MOD_enwiki_upos_skipgram_300_2_2021/cat_NOUN/)
* [Decrypto](https://www.decrypto.info/)
* [Matt Gaffney’s Weekly Crossword Contest](https://xwordcontest.com/)
* [This Is How You Lose the Time War](https://www.simonandschuster.com/books/This-Is-How-You-Lose-the-Time-War/Amal-El-Mohtar/9781534430990) by Amal El-Mohtar and Max Gladstone
* [Sinéad O’Connor – Troy (Live At The Dominion Theatre, 1988)](https://www.youtube.com/watch?v=1lV21J75vFE)
* [Get a Scriptnotes T-shirt!](https://cottonbureau.com/people/scriptnotes-podcast)
* [Check out the Inneresting Newsletter](https://inneresting.substack.com/)
* [Gift a Scriptnotes Subscription](https://scriptnotes.supportingcast.fm/gifts) or [treat yourself to a premium subscription!](https://scriptnotes.supportingcast.fm/)
* Craig Mazin on [Threads](https://www.threads.net/@clmazin) and [Instagram](https://www.instagram.com/clmazin/)
* John August on [Threads](https://www.threads.net/@johnaugust), [Instagram](https://www.instagram.com/johnaugust/?hl=en) and [Twitter](https://twitter.com/johnaugust)
* [John on Mastodon](https://mastodon.art/@johnaugust)
* [Outro](http://johnaugust.com/2013/scriptnotes-the-outros) by Jake Weisblat ([send us yours!](http://johnaugust.com/2014/outros-needed))
* Scriptnotes is produced by Drew Marquardt and edited by [Matthew Chilelli](https://twitter.com/machelli).
Email us at ask@johnaugust.com
You can download the episode [here](http://traffic.libsyn.com/scriptnotes/607standard.mp3).