The script was excellent. I had worked on it for three weeks. Every sentence served a purpose. The opening line hooked the audience. The build was clean, each beat escalating naturally. The humor arose organically from the situation. The climax landed with precision. On paper, it was the best script I had ever written for a performance piece.
The first time I performed it on stage — a corporate event in Linz, about a hundred people — I could feel the disconnect within thirty seconds. The words were right, but they did not sound right. They sounded composed. Written. Crafted. They sounded like what they were: sentences that had been carefully constructed at a desk and then memorized.
The audience could tell. Not consciously — nobody was thinking “this sounds scripted.” But there was a subtle withdrawal of engagement that I recognized from my consulting career. It was the same withdrawal I used to see in boardrooms when a presenter shifted from genuinely explaining something to reciting prepared remarks. The body language changes. Eye contact becomes intermittent. The energy in the room drops a half degree. People do not disengage entirely, but they lean back instead of leaning forward.
I was delivering good material badly, and the delivery was undermining everything the material was designed to achieve.
The Writing-Speaking Gap
Pete McCabe addresses this directly in Scripting Magic, and when I read his treatment of the problem, I felt the specific relief of someone discovering that the thing they thought was a personal failing is actually a universal challenge.
The core issue is this: writing and speaking are fundamentally different modes of communication. Written language has a rhythm, a structure, and a vocabulary that sound natural on the page but artificial in the mouth. Spoken language is looser, more fragmented, more repetitive, more alive. When you write a script and then memorize it word for word, you end up speaking in written language. And written language, spoken aloud, sounds scripted. Always.
Think about the difference between a letter and a phone conversation. A letter is grammatically complete, structurally organized, and progresses logically from point to point. A phone conversation is full of interruptions, restarts, half-finished thoughts, tangents, and verbal tics. But the phone conversation feels real and the letter, read aloud, sounds stilted.
The same dynamic plays out on stage. A well-written script, delivered exactly as written, produces dialogue that is structurally perfect and emotionally dead. The audience hears complete sentences, proper grammar, and logical progression, and their subconscious interprets these as signs that the speaker is not being spontaneous. Because in everyday life, spontaneous speech does not work that way. People who are genuinely talking to you do not speak in perfectly formed paragraphs.
My First Failed Solution
My initial response to the scripty problem was to stop scripting. If the script was making me sound artificial, I reasoned, the solution was to improvise. Work from bullet points. Know the structure but find the words in the moment.
This was a disaster. Without a script, my patter devolved into rambling. I hit the key structural points but arrived at them through circuitous paths that cost me the audience’s attention. My word economy collapsed. What the script accomplished in three tight sentences, my improvisation accomplished in twelve loose ones. The transitions between phases of the routine became sloppy. Jokes that had been precisely worded on the page became approximately worded in the air, and the approximate versions were measurably less funny.
Worse, without the safety net of a script, I found myself falling back on stock phrases — the kind of generic magician patter that I had spent months deliberately eliminating. “Let me show you something interesting” crept back in. “Watch this” reappeared. The verbal tics that Weber catalogs in Maximum Entertainment — the “alrights” and “let’s try” and “you’re gonna love this” — came flooding back because they filled the gaps that the script had previously filled with purpose.
I was now natural-sounding but substantively empty. Which was worse than scripty-sounding but substantively rich. Neither extreme worked.
McCabe’s Middle Path
The solution, which McCabe lays out with practical clarity, is a middle path that I have come to think of as scripted spontaneity.
Write the full script. Write it carefully, edit it ruthlessly, refine it until every word earns its place. Then identify the key lines — the sentences that must be delivered exactly as written. The opening line. The laugh lines. The critical transitional phrases. The climax setup. The closing line. These are the load-bearing walls of the structure.
Memorize those key lines word for word. They are locked. Non-negotiable. They will sound slightly scripted, but that is acceptable because they are the moments of highest craft — the jokes, the reveals, the emotional beats — where precision matters more than naturalness.
For everything else — the connective tissue between the key lines — know the ideas, not the exact words. Know what you need to communicate in each section, but find the words fresh each time you perform. The script tells you what to say in those sections. Your mouth tells you how to say it, in real time, using whatever language feels natural in that specific moment with that specific audience.
The result is a performance that has a scripted backbone with conversational flesh. The architecture is intentional. The surface is organic. The key moments hit with precision. Everything between them breathes.
The Rehearsal Shift
Implementing this approach required me to completely change how I rehearsed. Previously, I had been rehearsing the way I memorized presentations in consulting: read the script, repeat it, repeat it again, repeat it until the words came automatically. This is memorization, and it produces exactly the robotic delivery I was trying to avoid.
The new rehearsal method works differently. First, I memorize the key lines in isolation. I drill them the way a musician drills a tricky passage — repetition, variation, repetition. I say them at different speeds, with different emphases, in different emotional registers. The goal is to know them so deeply that they feel like my own words rather than someone else’s words that I have memorized.
Then, for the connective sections, I rehearse by talking through the ideas out loud, differently each time. I stand in my hotel room and explain the concept to an imaginary audience, using whatever words come naturally. Then I do it again, differently. Then again. Each time, the language shifts slightly. The core ideas stay the same, but the expression is fresh. Over dozens of these run-throughs, certain phrasings start to recur because they work better than others. These become my natural way of expressing those ideas — not memorized, but habitual. Earned through repetition rather than imposed through memorization.
The distinction is subtle but critical. A memorized phrase and a habitual phrase sound completely different when spoken aloud. The memorized phrase has a certain flatness, a mechanical quality, a sense that the speaker is retrieving words from storage. The habitual phrase has life. It sounds discovered in the moment even though the speaker has said something very similar dozens of times before.
The Weber Connection: Ignore the Punctuation
There is a vocal technique from Weber that connects directly to solving the scripty problem: ignore conventional punctuation when speaking.
Written language pauses at commas and stops at periods. That is how it is supposed to work on the page. But when you pause at every comma and stop at every period while speaking, your delivery acquires a metronomic quality that signals “rehearsed text.” Real speech does not follow the rhythm of written sentences. Real speech pauses in unexpected places — in the middle of a phrase, after a conjunction, before an important word rather than after it.
When I started deliberately inserting pauses in counterintuitive places — breaking up my sentences differently than the punctuation suggested — the scripty quality diminished significantly. The same words, delivered with different pauses, sounded less like text and more like thought. The audience perceived me as thinking and speaking rather than reciting and delivering.
This technique alone — just rearranging the pauses — made a bigger difference than I expected. It taught me that the scripty problem is as much about rhythm as it is about word choice. A performance can have the exact right words and still sound scripted if the rhythm is the rhythm of written prose. Change the rhythm to the rhythm of spoken thought, and the same words come alive.
The Conversational Checkpoint
I developed a self-diagnostic test that I now apply to every routine before it goes into the show. I call it the conversational checkpoint.
After rehearsing a routine to the point where I am comfortable with it, I find a friend or colleague — someone outside of magic, someone who has no context for what I am working on — and I explain the routine to them as a conversation. Not perform it. Explain it. “So there is this moment where I ask someone to think of something, and then I…” I describe what happens, why it matters, why the audience should care.
While I am doing this, I pay close attention to how I am talking. The language I use in conversation — the natural phrasing, the casual rhythm, the shorthand, the digressions — is the language my performance should sound like. If there is a significant gap between how I describe the routine in conversation and how I deliver it on stage, the stage version is too scripted.
This does not mean the stage version should be identical to the conversational version. Performance has its own demands — projection, precision, timing — that conversation does not. But the gap should be narrow. The audience should feel like they are being talked to, not performed at. The conversational checkpoint reveals whether that is happening.
Where I Am Now
The scripty problem never goes away entirely. Every time I develop new material, the early performances have a slightly written quality that takes ten or fifteen shows to wear off. That wearing-off process is the material making the journey from page to voice — from crafted language to owned language.
What has changed is that I no longer panic about it. I know the trajectory. The first few performances will feel slightly stiff. The key lines will be solid because they were drilled in isolation. The connective tissue will be slightly formal because I have not yet found my natural way of expressing those ideas in real time. And then, performance by performance, the language will loosen. The pauses will migrate to their natural positions. The words will stop being retrieved and start being spoken.
The script gives me the structure. The performances give me the voice.
If I had to distill the lesson into a single principle, it would be this: write like a writer, but speak like a human. The quality of the writing determines the quality of the ideas, the structure, and the key moments. The quality of the speaking determines whether the audience experiences those ideas as genuine communication or as theatrical recitation.
The words on the page are the blueprint. The words in the air are the building. They should share the same structure but feel nothing alike.