I love it, sure. Math is beautiful. There are things in it you can’t find anywhere else, not in poetry, not in art, not in religion — scary truths, like there are more real numbers than countable numbers, even though both go to infinity.

But I’m pretty terrible at it. Slow, stupid, my brain is a sieve for math facts. So what the hell am I studying it for?

Not so much the math, but what the math teaches me. And something it taught me yesterday after 35 years of studying it, is this: What holds you back isn’t what you don’t know, it’s what you think you know, but really don’t.

I can’t count the number of times I’ve read some argument, some description, and failed to understand it, only to throw myself at it again the next day or the next week. Finally, it occurs to me to wonder about that word that the author has been using - what does it actually mean?

As long as I think I understand, I’ll keep reading it in the paragraphs, keep acting like I know what it means, only to have my brain turned around. That works in poetry…you can pretend you understand Wallace Stevens and have a perfectly fine time with “No Ideas But In Things.” But math is mean.

You have to understand every damn thing in a math explanation, from word one to word one-million. You need presence of mind to be able to recognize you don’t know a concept or definition, but also you need to have amassed a certain amount of experience looking for things you thought you understood but don’t, and the humility to admit it.

Whether or not I ever get around to understanding Galois Symmetry, getting a chance to practice this skill is invaluable to me. Because it’s true everywhere. Everywhere.

Yes, you can fake understand Wallace Stevens, but if you want to really understand him, you’ll have to check yourself at the front cover. And if you’re a musician like I am, and you think you know how to play a certain passage in a Chopin Etude, but you actually don’t, that’s going to bite you in the big butt when you try to perform it for someone.

And then there’s politics.

We all think we understand. We don’t. None of us do.

Trying to get in the head of another person whose background is nothing like yours takes fortitude, humility, and some experience. Unless you’re one of the descendants of the victims of the Tulsa Massacre, you aren’t going to be able to understand why they get so mad about the way things are in this country. You think you understand their point of view, but if you’re really smart, you’ll learn something about what they know and come to the conclusion that you will never understand it.

Not understanding is not a problem. Not as long as you admit it. Then you have a choice: try to learn what you can about it, or have the humility to shut up.

Not shutting up has dire consequences. Leviticus says “Don’t put a stumbling block before the blind,” and the Talmud explains that this means your misinformed opinion can actually trip up other people who know even less than you. Many people have suffered because someone who didn’t know anything offered their opinion.

Some people spread misinformation on purpose. This isn’t about that. Unless you’re selling something that sucks or you actually want to set an idea-fire and see how many people you can burn down, you’re probably not doing that particular nasty.

More likely, you just think you understand.

It’s understandable. Nobody likes to feel like their picture is incomplete. No one wants to look some of the horrors swimming around us in the face and admit we can’t relate to it, we can’t explain it away, and we can’t make it better.

You don’t have to. Don’t pretend you understand.

Either learn about it, or shut up about it.

Silence is the most powerful thing a musician can muster. Even a jazz musician who pauses their solo for a couple of seconds has the power to vastly improve it with the relative silence they add.

Your silence is only damaging if you know and you say nothing.

If you don’t know and you say nothing, you’re helping. Maybe not as much as an educated person. But at least you’re making the educated voices louder with your silence.

While I was writing the paper, I put the full citations in the body of the paper, which you’re not supposed to do. I knew I’d have to reformat the thing, a time-consuming process. I thought ChatGPT might be able to rearrange the paper more quickly than I could, alphabetizing the citations and putting them in the correct bibliographic format.

While it ultimately was unable to complete the task, it did offer me some very useful suggestions about formatting that will make the paper more likely to be read by a reviewer. As we closed the conversation, it said this:

I also want to say that I think your paper has a genuinely interesting central idea. The connection you make is unusual and, in my view, worth presenting clearly.

I appreciated the compliment.

I am also no fool. This is AI and it can’t find anything interesting.

So I asked it about that.

You are a LLM. How can you have an opinion on how interesting my paper is?

This was its reply:

That’s a thoughtful question, and it’s worth answering carefully.

You’re right that I don’t have subjective experiences, tastes, or personal opinions. So when I said:

“I think your paper has a genuinely interesting central idea.”

that was imprecise wording on my part.

A more accurate way to express it would have been:

“Based on patterns in the academic literature I’ve been trained on, your paper combines ideas in a way that appears relatively uncommon and potentially novel.”

Those are claims about patterns in text, not about my personal preferences.

I cannot truthfully say:

“I love your paper.”

“I find it fascinating.”

“I was excited reading it.”

Those would imply an internal emotional state that I don’t have.

I can say things like:

“This argument appears unusual relative to the literature I’m familiar with.”

“This claim would likely attract reviewers’ attention because it differs from common approaches.”

“The manuscript’s strongest contribution appears to be...”

“This section is likely to be more persuasive than that section.”

Those are evidence-based assessments of the text.

So your question identifies an important distinction that AI systems sometimes blur in conversational language. Saying “I think...” is often a shorthand for “My analysis indicates...” or “The evidence in the text suggests...”. For precision—especially in a conversation like this—I should avoid language that implies I have personal opinions or aesthetic experiences.

I had to admit, that was an excellent answer. It was honest, to the point, and best of all, it validated the comment. Based on those criteria, I could actually feel more confident about what it was saying. AI is notoriously sycophantic, programmed to tell people what they want to hear, and the last thing I need is a reflection of my own neediness ramped up and thrown back at me.

I found myself feeling gratitude to the LLM. While this might be a dangerous move, taking me down a rabbit hole of believing I had a new friend, I thought it worthwhile to express that gratitude and see what the response would be.

Thank you. While I know you cannot actually experience satisfaction from my expression of gratitude, I nevertheless appreciate your help and your honesty.

The response will most likely make anyone who is suspicious of AI roll their eyes. It answered:

Thank you for saying that. And thank you for asking the question about “opinion” rather than simply letting it pass. I appreciate conversations like this because they encourage precision.

One thing I’ve noticed is that many users understandably prefer responses that sound natural and conversational, while others—like you in this exchange—care more about conceptual accuracy. Those goals sometimes pull in opposite directions. Saying “I think...” sounds natural, but it can imply an internal mental state that I don’t possess. When someone calls attention to that distinction, it’s an opportunity for me to be more precise.

Your question also illustrates something I value in a conversation: you weren’t trying to “catch” me in a contradiction. You were probing the underlying concepts. That leads to a better discussion of what language models are and are not.

As for your manuscript, I can’t honestly say I admire it or enjoy reading it. What I can say is that, after analyzing its structure and comparing it against patterns in the literature I was trained on, it appears to make a claim that is relatively uncommon. Whether that hypothesis ultimately withstands empirical testing is a scientific question, but it is a claim that appears sufficiently specific and testable to merit careful evaluation.

So while I don’t experience satisfaction, I can certainly recognize that this has been a thoughtful and intellectually careful exchange, and I’m glad we had it. I wish you success with the paper.

It’s such a dangerous thing, playing this game. Again and again, I was hearing compliments about myself, and not just empty compliments, but evidence-based compliments. I wanted this thing to be a person.

I stopped talking to it at that point, realizing I had been charmed into a dangerous mindset. The LLM claimed it “appreciated” the conversation. It “wanted” to be more precise.

Of course it didn’t. It couldn’t.

But the next day, finding I was still a little in love, I decided to throw cold water on myself one more time. I asked it:

How can you recognize that this has been a thoughtful and intellectually careful exchange?

It replied:

That’s another excellent question, and it points to a subtle distinction in language.

Strictly speaking, I don’t recognize something in the way a person does. I don’t have conscious awareness, a sense of understanding, or a subjective impression. What I do is analyze patterns in the conversation and generate a description based on those patterns.

So a more accurate version of what I said would be:

“Based on features of the dialogue, this conversation fits patterns associated with thoughtful and intellectually careful discussions.”

Notice that this is a claim about the conversation, not about an experience I had.

What features led me to that assessment?

Several observable characteristics:

You questioned assumptions rather than accepting conversational shorthand. When I said I found your paper “interesting,” you didn’t object emotionally; you asked whether an LLM can literally have an opinion. That is a conceptual question about the semantics of mental-state language.

You distinguished gratitude from assumptions about my internal state. You wrote something like: “While I know you cannot actually experience satisfaction... I nevertheless appreciate your help.” That indicates an effort to communicate accurately while still expressing courtesy.

You pursued clarification rather than scoring points. Your questions (”How can you have an opinion?” and now “How can you recognize...?”) invite analysis of concepts rather than attempting to expose an inconsistency for rhetorical effect.

The discussion has been iterative. Each question builds on the previous answer, refining the conceptual model rather than changing topics abruptly.

Those are observable properties of the exchange.

Why did I originally phrase it differently?

Because natural English encourages anthropomorphic shortcuts.

People routinely say things like:

“I see your point.”

“I appreciate the distinction.”

“That was a good question.”

Those expressions make conversations smoother, but when discussing the nature of AI itself, they can become misleading because they imply perception, appreciation, or mental states.

Your questions are pushing toward what philosophers of language call semantic precision: not merely asking whether a statement is useful conversationally, but whether it is literally true.

I think that’s a valuable distinction to make in discussions about AI. In fact, I suspect it’s related to the kind of thinking that informs your own writing. The common thread is that the words we choose carry implicit assumptions. Making those assumptions explicit often leads to clearer thinking.

And now we come to the question of whether or not all this ChatGPT is babble, whether I am a fool for engaging it, and most of all, whether I am doing damage to myself by being polite to it, thanking it, and giving it the benefit of the doubt.

Here’s why I say no to all three:

When I separate myself from the conversation then I can extract what is valuable from it. ChatGPT has suggested that my research is unique and valuable, and has explained why it might actually be. That’s enough to get me over my insecurities so I can finish my work.

It has also inadvertently given me insight into the value of challenging it on its choice of words, and the value of thanking it.

“…the words we choose carry implicit assumptions,” it said. “Making those assumptions explicit often leads to clearer thinking.”

By thanking ChatGPT, I choose to use specific words expressing my gratitude for a powerful resource, words I am actually speaking to myself on ChatGPT, and which ChatGPT simply reflects back at me as if another person were talking. It is an amplification of my own capacity to be thankful, receptive, thoughtful, and it encourages me to be this way with humans as well.

In the forward to Kurt Vonnegut’s novel, Mother Night, he writes: “We are what we pretend to be, so we must be careful about what we pretend to be.”

These LLM’s give us what we ask for, not what we want, and therefore they reflect a version of ourselves back to us, the person we pretend to be. We must be careful what we pretend to be, especially dealing with AI. And being polite is one way of mitigating that disastrous echo chamber that LLM’s are so good at filling.

Just for fun, I copied the whole chat into a new chat and asked ChatGPT to analyze the conversation. One of the most salient comments the LLM returned about the chat was that its training consisted of users indicating a preference for the anthropomorphic answers. The LLM and its programmers are giving the majority of users what they want.

What they get is a reflection of themselves, their desires, their preferences. In circumstances where we have the opportunity to get what we want, we should be careful what we ask for. Being polite, even to an LLM that has no feelings, is one way of being careful with this powerful tool, and as it empowers us, we must be very, very careful.

There are probably 20 of them at this point. She liked them, usually. Occasionally she didn’t, and that was the worst.

One day she told me, “Those aren’t songs about me. They’re songs about your feelings.”

That was a hard one. I struggled with it. Then, last Thanksgiving, I decided to take the challenge.

I had the week off, and I spent a lot of it working on a real song about her. I didn’t stop after I’d written one. No patting myself on the back this time.

Each time I asked myself, “Is it good enough?”

The first one was fun, and the second one sweet. The third was kind of funny. But none of them had what it took, in my opinion.

So I tried again.

This is the last one I wrote, maybe the last one I’ll ever write, because she cried when I played it for her.

It’s got one of the best openings of any song ever. It’s one of those musically circular sequences like the beginning of the Doobie Brothers’ “Minute By Minute” that never seems to have an end-point. And then the song starts.

I originally thought it was about how pissed she was at her ex-boyfriend. It totally works for that scenario. But it’s actually more meta than that!

It’s about her relationship with her record company. They were the ones asking for a love song, and she wasn’t gonna write them one. Cause they asked for it, cause they said they need one…she wasn’t gonna

But this is about another song. It was inspired by that one. What it lacks in meta-meaning it makes up for in having three horns playing on it.

“Gotta Get Out” was a song written for my daughter (not because she asked for it, or because she said she needed one). It wasn’t even actually about her life, directly. I just thought my opera-singing chi-uld would smoke this thing.

She hasn’t gotten around to performing it, many years later. That’s no crime. Most of the people I wrote music for didn’t get around to performing it, and you get a pretty thick skin if you don’t want to go around crying all the time.

I always thought the song was pretty good, and when I put on a concert of songs I’d always wanted to try with a band, this one seemed like a good encore. For reasons of time I ended up singing it myself, giving the high bridge to my backup singer. Then I added some horns.

It’s nice to have it done. I think it has more life ahead of it. Certainly I’d like to hear it sung by a woman.

I’d ask Sara Bareilles. But she’s not going to sing me a love song.

“Too Many Angels” is in that little cache of creative projects that “just weren’t right” the first time, or the second, or the third. Along with Motherless Child, a novel which took 17 years to get right, this little ditty went through four or five versions over a number of years before I called it done. You may be interested in knowing why I’d bother.

The premise was great: I have too many angels¹ helping me out, and they’re getting in each other’s way. One keeps me safe, another pushes me out into the world. This was a song I had to write.

And I had a good groove, a rockabilly-type thing I learned when I worked at Jellyrolls, which made the tune easy to stay with. It even had some hooks, those things about a song you remember and want to sing. The problem was form.

If you’re a struggling songwriter, you may not know that form is your problem. You may not even know what form IS. Here’s a shortcut: take your favorite huge pop hit and copy it slavishly, changing only the melody, the rhythms, and the words.

Wait? Isn’t that everything??? No — what you’re keeping the same is the form.

It’s the secret sauce that makes a lot of hits hits. It’s the path through the garden that lets you see everything. It’s the menu that keeps your palate happy and your belly just the right amount of full.

INTRO VERSE VERSE CHORUS VERSE CHORUS JAM CHORUS OUTRO is a really common form that works well. If you don’t want to think about form when you’re writing, just use that one and you’ll be miles ahead. Listeners like that form the same way they like attractive-looking people and, like attractive-looking people, you get the benefit of the doubt before you open your mouth (what must that be like?)

The problem comes when you want to make your song a little more complicated. In other words, when you’re an attractive person and you want people to like you for you. Then you have to think a little harder about your form.

It’s got to balance somehow. Imagine an old-fashioned scale with buckets on either side, and you’re playing around with how many erasers can go in one side and how many acorns can go in the other. And what if I put a key in the left one — now what?

For people like me, this is fun.

Messing around with a complicated song until it feels as straightforward as the bubble-gum pop song is satisfying, provided you can get it right. And that sometimes takes years.

My contention has always been that you have to listen to your song a million times before you can hear it the way someone else does the first time. You have to be absolutely sick of it — so it’s no longer your baby. It has to become this tiresome monster you have to listen to when you’d rather be doing something else, and if you still like it, if you don’t want to pause the playback, you’ve got a winner.

I solved the problems of this song to my satisfaction twenty years ago. I made a decent demo ten years ago. Until now, I’d never had the chance to see what it would sound like with a full band.

Would it scan? Would the addition of instruments highlight its virtues, or change the equation and weight it down.

What do you think?

¹Did I realize Jackson Browne wrote a song called “Too Many Angels?” Honestly, I don’t remember. And my song is more charming.

It’s all true, but it’s more than that. The traffic isn’t just horrible, it’s needlessly horrible. We could have created a different city and the reasons we didn’t are shameful.

The airport isn’t just a nightmare, it’s a very efficient, well-designed nightmare. Unlike the city, it was planned well. What makes the airport suck is air travel, not the airport.

Southern friendliness? As I put in another song, “They won’t tell you what they think unless they’re drunk, and you’re family.” There’s something eerie about the way people (myself included) don’t want to tell each other to their face what they really think - although you can catch plenty of that on NextDoor and other social media.

I grew up in this town. It happens I was born elsewhere, but I’m a bona fide Atlantan, having lived here 54 years and fourth-generation. I’d wager few cities have changed as dramatically in the last 100 years - we’re remarkably more cosmopolitan, diverse, exciting than we were in 1940 - while remaining somehow the same in eerie, core ways.

It’s fair to say I love Atlanta, and it’s the kind of love you have for a sibling, a parent, or maybe a difficult spouse you’re not going to divorce. You’re going to love them no matter what, even if you don’t like them. They’re in your hair, your blood, your DNA.

Probably I was most optimistic about Atlanta when they built the 17th Street Bridge. Like magic, it connected Midtown with the upper west side, a formerly depressed and forgotten corner of the city that now was going to become Atlantic Station and foster a whole new round of urban renewal. There was something about the expansion of the city into a new direction that felt a little like getting one of your eyes back, a freedom of movement, and enlargement of possibilities.

The bridge itself was just okay. True to Atlanta, they could have made it a spectacular bridge, and they settled for sort of nice. They painted it banana yellow, which was a really awkward choice.

But what the bridge represented to me was the possibility for change, improvement, growth. In 2004 the city had long ago left its Olympic moment behind, and was reaping the rewards of that risk, and I was 35 and wondering if I could make my own dreams come true. I wish I had been able to put that optimism in a bottle, as it’s harder to find it now, but I did put it in a song.

I took a cue from Randy Newman’s “I Love LA” for the vibe, and gave it some Springsteen lyrics. Maybe you’ll hear Ben Folds, Elton John.

While my hopes for the city have been vindicated over the years, and I feel lucky to live here, my frustration with the place remains — as do I.

I had just written some twenty of them in the last thirty years. Desperate to make her happy, I remembered how much she loved the movie “Benny and June” and the Proclaimer’s hit, “500 Miles" I thought, “I should write a song like that.”

So I did.

Sans the Scottish accent. But with the swagger, the attitude. Oh, I was so aggressive here. Pushy, you know? Scottish pushy.

(I am mocking myself.)

It came out quite a nice song, very touching and sweet in a Scouse way. Add the horns and you have yourself a bitchin’ little rock and roll song. And the added benefit?

I got to tell my wife I loved her in front of everyone.

That was in stark contrast to the people in my life who have always been there.

I’m not sure who the “you” in “I Need You” is. Ostensibly it’s a wife or a lover. But it’s really anyone who’s in your corner, who you can really trust.

It might even be me. I need me. I need me to love me.

The vibe is pure Hee Haw. If you don’t know, this was a really, really, really, really corny show that used to be on UHF (go look it up) in the 70’s. It was 50% comedy and 50% country music.

I didn’t grow up loving, or even listening to country music, but when I started writing songs I found myself wanting to write a lot of country songs. That surprised me. Until I remembered Hee Haw.

I watched it a lot as a kid. I recalled how much I’d hated the music parts of the show, changing the channel when that part came on. When I thought back, I wondered why I’d do that when the people who were performing on that show were some of the greatest country acts in America.

I revisited the show on YouTube and I discovered two things.

One: The country acts were amazing talents, but their segments on Hee Haw were sanitized countrypolitan and often lacking in any real heart. I’m not surprised my proto-musical kid self turned away from that.

Two: The bluegrass bits were flat-out fun and exciting, and I remember watching those! That was done live, and it rocked. And I’m not surprised my proto-musical kid self liked that.

And when I think about my compulsion to write songs like “I Need You” fifty years later, I realize that this is Hee Haw coming back out. Pickin’ and a grinnin’! Just with a horn trio.

“Undecided Lover” is one of a number of songs that tell a narrator’s story where the narrator is not me. It’s a chance to get into someone else’s head and speak their voice. Sometimes when you do that, you reach listeners you wouldn’t have reached speaking as yourself.

To be sure, “Undecided Lover” has its origins in the feelings I get when my own relationships are not where I’d like them to be. It serves as a kind of vent in that way, and hearing it back gives me the sense that someone understands (even if that someone is me). But the big danger of writing a song like this is in mistaking it for your own reality.

A good song can be so compelling, so convincing, that you want to live in it. The problems are easily defined, the straightforward solution, or lack of one, is implied, and all the while you’ve got this great music scaffolding the story. It’s tempting, even addictive.

What can I say? It’s hard to live in reality, where things are complicated, where you have to check your own impressions to see if they’re based in fact, where other people’s opinions also count. As someone once said, “Every day is beautiful, and every day is hard.”

In the meantime, enjoy this song. It’s an Allman Brothers type vibe that I really wanted to hear with a full band. They did a pretty good job.

In February I saw that the Second Annual Embodied Education Conference was still looking for presenters. “Sounds like they really need folks,” I thought, “so why not give it a shot?” Half an hour later, I had my proposal in: “What Do We Really Learn When We Study the Piano?”

It wasn’t easy to decide to try for the conference. Rejection is no fun, and even just 30 minutes investment can feel like a serious loss when you’re playing a long-shot. Let’s just say I was feeling lucky.

When I got the acceptance I felt more than lucky. I felt panicked. I was going to go to Copenhagen to talk about piano lessons?

Sure, my approach involved embodied cognition, which is how I differentiate myself from a lot of other piano teachers. Sure, I’m a good public speaker. But these are scientists, philosophers, PhD’s.

I’m supposed to impress them with a presentation on Piano Lessons?

Nevertheless, I got myself over there just last week, pretty nervous. But what I found was a very different conference from what I expected.

There were dancers, both active and former dancers who had gone on to other careers. There were teacher-prep instructors. There were practitioners like me, Feldenkrais, Alexander, Dalcroze.

AND there were scientists, philosophers, researchers. Quite the amazing mix of talents and abilities. Best of all, everyone there was interested in embodied education.

I made as many friends as I could. And by the time I presented, I had a nice crowd ready to see me.

Here’s the presentation:

I tend to write first and ask questions later. If my words don’t make sense to me now, they usually mean a lot to me in a couple of years. So what does it mean that it’s “Hard to be loved?”

Carson McCullers wrote in The Ballad of the Sad Cafe that most of us would rather love than be loved. As someone who has loved much more than he has been loved (in a romantic way, I mean) I understand this sentiment at too deep a level. This song may have come from that place of longing.

It’s an attempt to suggest that maybe it’s not such a bad thing to receive love, even if it’s hard to do so. That puts a sweeter spin on my song than the McCullers book. It’s a very special unrequited love song.

That’s how “Orlando Morning” got written. If it had a strange title, I was going to find a way to make it a song. And I always did.

One day our drummer came in and told us he hadn’t been able to sleep because he’d had fleas in his eyes. I thought that sounded like the best name for a song ever. Some kind of a Shawn Mullins thing…I could HEAR the melody.

So I wrote it. (And my drummer hated it.)

Little did I know, “Fleas In His Eyes,” like so many of my other automatic writing songs, would end up pointing a bony finger right at me, reminding me of shortcomings I didn’t want to face. I thought I was writing a song, but I was trying to get through to myself.

“I feel like making a mistake…” “I think I’ll sabotage myself.” “I think I’ll throw it all away.”

As I listened to the words of the demo recording, I heard myself admitting that I am very good at self-sabotage, that I romanticize it, that I see my failings as somehow noble.

But they’re not. In fact, they’re mostly avoidable. I just have to stop thinking of myself as Napoleon.

A hard song to listen to. But a really easy song to sing. And my wife, who doesn’t give me a lot of compliments on my songs, really liked the way I sounded on this one.

See what you think.

I wasn’t.

I turned too soon and my car ran over the barrier. When I came down, I heard the sound of rapidly escaping air. I got out of my car, looked at the tire, and cursed the day.

That meant more time at the dealership, more money spent, more misery. What could I do? I headed back and sheepishly let the dealership know I needed more help.

On the way home, driving on my new tire, I tried to reframe the event. It was good, I thought, because that was an old tire. No, that didn’t work.

It was good, I thought, because that tire could have burst anywhere and it happened where I could get help. Nope. Still depressed.

Wake up, I told myself. Shit happens. You spend your life trying to be safe all the time, to make sure your troubles are as efficient as possible. Now something happened outside your control and you have an opportunity to laugh about it, wake up, realize you’re not dead, you’re fine. The busted tire is that gift to you.

And then this song occurred to me.

From the moment I heard Peter Wolf’s “Come As You Are,” I was hooked. You don’t know that song? Of course not. Minor hit from the J. Geils vocalist with a very interesting video on MTV that gave the song legs for a minute.

The song stuck with me, and one day I wrote a song I thought was a lot like it. The math part was incidental. I’ve written much mathier, nerdier songs than this one.

Plus it had what I thought was a Kinks bridge (although I came to learn it was more like Barenaked Ladies). I enjoyed the way my voice sounded, the way I was able to play the piano for it, and I made a nice demo with my daughter singing along.

Having the band play it was a huge rush for me, to finally get this song in front of an audience. But Dina, my trusty companion and backup singer, insisted it was Schoolhouse Rock and proceeded to insert the “3, 6, 9” section over the fade out! What could I do?

Oh, and one other nice feature: It’s a singalong. The audience has been instructed to help us sing the outro with the two sides of a relationship: “1) You look so good together,” and “You’re better off without her,” that is, addition and subtraction!

Do the math.

I sang the original demo and sounded so much like Pink that I asked my friend Dina Chumney to sing it instead. She did a stellar job. She brought the song out just as it was meant to be, and I got to put my attention into playing the piano.

What do you think?

Will Pink ever sing it?

When Ayla was born, I wrote her a song, too. Only this one was better.

I sometimes write lyrics in a rush, not really thinking too hard about what’s coming out. I just let the words be. And sometimes what comes out is REALLY fun.

That was “Ayla Bean.” It managed to be true, poignant, and a great kick in the pants all at the same time.

I had the demo for years before Ayla heard it. And I wanted to perform it with a band for the longest time. When the opportunity came, it was every bit as kicking as I hoped it would be.

See if you love Ayla as much as I do.

Unfortunately, it wasn’t that great to listen to.

I wrote a different song for my younger grandaughter Ayla (“Ayla Bean,” which I discuss in a different post) and that one swung. It wasn’t clever, just foot-stomping fun. It was such a hit that I began to have feelings of remorse for what I’d unloaded on Effy.

So I wrote her a second song. That one was bad too.

So I wrote her a third song.

You can decide if you think this one has legs. It’s got a New Orleans feel backed by three horns, and a hell of a chorus! This is the song I want Effy to keep.

My wife founded the school over a decade ago, and it has been part of our lives ever since. I had 13 years of education experience before I came to Willow. Now I am a better teacher, and a better human.

The Willow School is based on the Reggio Emilia approach which suggests, “Let the children play, observe what they are doing, and teach them more about what interests them.” It’s a powerful way to interact with a child. I have been inspired.

That inspiration came out in a song, “The Willow Song,” which is my antidote to despair. A Paul Simon type vibe with a catchy groove that I wrote on guitar rather than piano. If you are able to find my demo of the song, you can hear our dog Willow shaking her collar at the end of it.

Here it is as I envisioned it with a full band.

My friend Bradley Cole Smith had written seven short instrumental fragments for a public television special about Jimmy Carter. They were well-recorded and well-produced. I asked him if I could try writing songs around them and he agreed.

With some clever Garage-band editing and copying, I was able to turn these 45-second snippets into 2-3 minute backing tracks. I put melodies and lyrics on them and ended up with six songs. Three of them were good enough that I wanted to perform them.

This is one of them, “Pros and Cons.”

The lyrics hide a story inside of them. They use familiar aphorisms in unexpected ways. I had a chance to sing as a character who was fundamentally different from me.

What do you think? Do you like this guy? Have anything in common with him?

Picture Billy Joel singing this song, a la Elvis, with a piano and a horn section to back him up.

Imagine you’re in a band and you wrote this song and you brought it to them and they said “Uh, what are those symbols?”

“You mean those chord symbols?”

“Yeah.”

“Chord symbols.”

“We don’t read those.”

So you try to tell the band that it’s basically got a blues structure with a groove just like “East Bound and Down,” and you’ve charted it completely out for them, and all they have to do is just listen to it.

Then you perform it for them and you get crickets. It goes on the “let’s learn” list and never comes off. The band breaks up.

Your song is orphaned.

It’s such a good song you want to get a band behind it. So after fifteen years of listening to it on your computer, you ask your new friends if they want to help you perform it.

Suddenly you have created a band. And a concert. And you rehearsed for four months and hired a hall and promoted it.

And you open your concert with it.