Working With The To Have And Have Not Screenplay
The 1944 screenplay for To Have and Have Not is one of those documents that looks straightforward on the surface but falls apart fast if you try to use it as a primary reference for anything serious. I ran into this when a director on a low-budget period piece asked me to track down the original shooting script because the published book version had been heavily altered. What I found was a mess of typist errors, studio-mandated rewrites, and several sequences that never made it to the final cut. If you're looking for the To Have And Have Not Screenplay for research or production purposes, you need to understand which version you're actually getting. The most common version floating around is the 1945 published edition from Random House, which was adapted by Howard Koch and the Wachskos (Murray and Richard) from the Ernest Hemingway short stories. That's not the shooting script. That's not even particularly close to it. The actual production documents live scattered across the Warner Bros. archives, the Academy Screenplay Collection, and various private collector holdings.
Locating The To Have And Have Not Screenplay
There are three main versions you will encounter, and they differ significantly enough that mixing them up causes real problems on set or in academic work. The first is the preliminary draft by Jules Furthman, written before Koch came on board. Furthman's version is substantially different in structure. The character of Harry Morgan operates more like a straight smuggler throughout, and the Marie wave scene plays out quite differently. If you're doing dialect work or character analysis, this draft matters because it shows where the studio steered the tone toward romance and away from the harder-edged Hemingway source material. The second is the Koch shooting script, which is what actually went into production. This is held at the Margaret Herrick Library. You can request access if you have a legitimate research need and proper credentials. The script runs approximately 112 pages in the standard industry format with the usual production notations - page revisions in different colored paper, scene numbers added in pencil, and notes about location adjustments for the Bahamas shoot.
The third is the published Random House edition from 1945. This is the one you can actually buy and read right now. It's decent for general reading but contains dialogue modifications that Koch made after principal photography wrapped. Some lines were tightened for the Marine Midget series re-release in 1952, and those changes bled into later printings. If you cite the Random House edition in an academic paper without noting which printing, reviewers will catch it. I once spent three weeks trying to reconcile a scene where Steve comes to claim Marie at the end. The published script has a clean exit. The shooting script has an extended exchange that was trimmed during post-production because the pacing dragged. The Furthman draft has yet another version. The difference isn't cosmetic. It changes how you read Harry's final choice - whether he lets her go willingly or whether there's a lingering threat underneath the farewell. For a production reimagining, this distinction is the difference between playing it safe and actually saying something.
Get the Full Details

What This Screenplay Actually Shows You
The thing most people miss about To Have and Have Not is how aggressively it departs from Hemingway's original story. The short story "To Have and Have Not" is a bleak, nearly plotless meditation on a man who sinks into moral compromise out of sheer necessity. The screenplay turns it into a wartime adventure romance with a Central American revolution subplot tacked on. Koch and the Wachskos did this deliberately. The studio wanted a vehicle for Bogart and Bacall, and Hemingway's story was too downbeat for that purpose. The revolution subplot, borrowed from "The Snows of Kilimanjaro" and other Hemingway material, is the most obvious sign of assembly-line writing. It gives Harry a reason to be involved in international intrigue and provides a natural excuse for the action set pieces. But it also creates a structural problem. Harry's motivation shifts halfway through from personal survival to ideological commitment, and the screenplay never fully reconciles the two. You can see this in the second act where the tone pivots from smuggling comedy to wartime urgency without much transitional work. From a craft perspective, the real interest lies in how Koch handles the dialogue between Harry and Marie. The famous banter - the stuff that launched a thousand copycat scripts - was almost entirely Koch's invention. Hemingway didn't write dialogue like that. He wrote in sparse, declarative sentences with subtext buried in what characters don't say. Koch filled the space with wit and sexual tension, which is why the script feels so different from the source stories even though it carries the same title and characters.
If you're studying this screenplay to learn about romantic banter or wartime genre conventions, it's valuable. If you're using it as proof of how Hemingway wrote, you're going in the wrong direction entirely. The gap between Hemingway's prose style and Koch's script dialogue is enormous and instructive. Koch understood that film dialogue needed rhythm and entertainment value in a way that literary realism doesn't always demand.
Practical Problems With Using This Material
The biggest issue I've run into repeatedly is that many online PDFs of the screenplay are either scanned copies of the published book or fan-made transcripts that contain errors. A scanned Random House edition will have page numbers from the book layout, not the original screenplay format. That means scene headings won't align with industry standards, and you'll waste time translating between the two systems. Fan transcripts are worse. I found one circulating on a screenwriting forum that was clearly typed from memory with scenes in wrong order and dialogue attributed to the wrong characters. It cited the source as the "original shooting script," which made it sound authoritative until you actually checked it against anything else. Always verify against at least two sources before trusting a transcript. Another problem is that the Warner Bros. archives don't have a single complete document. You get drafts, continuity sheets, production notes, and final shooting scripts spread across multiple boxes. If you're doing serious scholarship, plan to spend time at the Herrick Library rather than relying on digital scans. The physical documents contain marginalia and production decisions that disappear in reproduction.

There's also a copyright issue worth noting. The screenplay itself is under copyright held by Warner Bros., which means you can't reproduce substantial portions of it without permission. Fair use covers analysis and quotation in academic or critical contexts, but if you're planning to publish a full transcript or build a commercial product around it, you need to clear rights with the studio. I learned this the hard way when a producer I consulted with got a cease-and-desist for including the full script in a development package.
What To Do Instead
If your goal is simply to read the screenplay, the Random House edition is accessible and serviceable. Order it used from AbeBooks or find a library copy. Don't pay more than twenty dollars for it. If you need the shooting script for professional work, apply for access at the Margaret Herrick Library. You'll need to show a valid research purpose and agree to their viewing rules, but they're generally reasonable about it. For students of screenwriting, the Furthman draft is freely available in some university archives and occasionally surfaces in rare book auctions. It's worth studying because it shows how a screenplay gets rewritten by committee. The differences between Furthman and Koch reveal where the studio pushed for changes and where the writers fought back. That process is more educational than either script alone. There's no single perfect version of the To Have and Have Not Screenplay sitting in one place. The material exists in fragments across published editions, archival collections, and secondary sources. The best approach is to treat each version as a separate document with its own context and limitations rather than assuming one of them represents the definitive text. That mindset will save you from the kind of errors that come from treating a published book as if it were a production document.