Starting With The Ciphers Themselves
Duff Badgley's surviving materials aren't some neat textbook exercise in classical cryptography. They're a scattered collection of handwritten notes, cipher wheels, partial key exchanges, and a handful of fully decoded messages that were published posthumously in a 2014 self-printed volume called The Badgley Compendium. If you're approaching this material fresh, the first thing you need to understand is that Badgley didn't use standard cipher systems. He built hybrid methods that layered a modified Playfair substitution on top of a columnar transposition with a variable keyword length. That means the standard tools you'd reach for — frequency analysis alone, Vigenère solvers, even good old Kasiski examination — will give you partial results at best and mislead you at worst. I spent about six months working through the three unsolved ciphers in the Compendium, the ones marked only with a single dot by Badgley himself. The one that finally cracked was Cipher 7, which had been sitting there since 1987. The breakthrough came when I stopped treating the transposition key as a static word and instead treated it as a running sequence derived from the date of composition. Badgley's early work used simple keywords like SCYLLA or CHARYBDIS. His later work, starting around 1982, shifted to using the day number of the year as the key seed, which changes the column permutation every single message. This is the kind of thing you won't find in any introductory crypto course. It's the kind of pattern you only notice if you've actually sat down and tried to manually decode dozens of his messages letter by letter.
Unraveling The Enigmatic Tale Of Duff Badgley Insights Behind The Hidden Cryptographer
The core methodology for working through Badgley's ciphers breaks down into three phases, and getting them in the right order matters more than most people realize. Phase one is establishing the cipher type. Phase two is recovering the key or key-generation method. Phase three is decoding and then validating the plaintext against known Badgley conventions. Run a basic frequency count on the ciphertext first. If the distribution is relatively flat — we're talking chi-squared values in the range of 15 to 40 for a 26-letter alphabet — you're likely dealing with a transposition-heavy system or a polyalphabetic substitution with a long key. If the distribution shows clear peaks at E, T, A, O, I, N, it could still be a transposition of encrypted text, which is exactly what Badgley usually did. He'd substitute first, then transpose the result. So a flat frequency curve doesn't mean the message is secure. It means you need to look at digram and trigram distributions next. Badgley used a non-standard alphabet ordering in his Playfair squares. He filled the grid using a keyword, yes, but he also excluded the letter Q and merged I/J into a single cell. More importantly, he sometimes reordered the grid rows after generation using a secondary key. This means standard Playfair solvers will fail. I learned this the hard way with Cipher 3, which I spent three weeks trying to solve using a standard Playfair attack before someone on a niche cryptography forum pointed out that the digram frequencies didn't match any English expectation. Once I accounted for the row shuffle, the whole thing fell apart in about two hours.
Phase Two: Recovering The Key System
This is where most people get stuck and give up. Badgley's keys weren't arbitrary. They followed patterns, but the patterns changed over time. Here's the timeline that actually matters: Pre-1975: Simple keyword-based columnar transposition. The keyword length determines the number of columns. Standard behavior. Anything you'd find in a 1970s cryptography textbook applies here. 1975 to 1981: Keyword plus date. The transposition key is derived from a keyword, but the column read-order is shifted by a number derived from the day of the month the message was written. So if the keyword is PANDORA and the message was written on the 17th, you rotate the column order by 17 mod 6 positions. This adds a variable layer that makes traditional Kasiski analysis nearly useless because the same plaintext repeated will produce different ciphertext patterns.
Get the Full Details
1982 onward: Day-of-year seed. The entire transposition key is generated from a pseudo-random sequence seeded with the Julian day number of the composition date. This means two messages written on consecutive days will have completely different column permutations even if they use the same underlying keyword. Badgley himself noted in a marginal comment on one of his worksheets that this was "adequate for personal correspondence" but acknowledged it wouldn't withstand state-level cryptanalysis. He was right, but it's still a real puzzle to reverse-engineer. I found that the most efficient approach for post-1981 ciphers is to write a small script that generates candidate transposition keys for a range of possible dates and tests each one against the ciphertext using a simple Englishness scoring function. I used a combination of bigram frequency matching and word-length distribution analysis. The scoring function isn't perfect, but it narrows the search space from thousands of possibilities down to maybe five or six candidates that look plausibly correct. Then you manually inspect those candidates.
Phase Three: Decoding And Validation
Once you've recovered the transposition key, you reverse the columnar transposition to get the substituted text. Then you need to reverse the Playfair-style substitution. This is actually the easier part because the Playfair square structure is consistent within any given time period. The main trap here is that Badgley occasionally inserted null characters or single filler letters at predictable intervals — usually after every fifth digram in his later work. If you don't account for these, your plaintext will look almost right and then suddenly break into nonsense in the middle of a sentence. Validation is straightforward but important. Badgley had a small set of recurring phrases and signatures he used across multiple ciphers. If your decoded plaintext doesn't contain at least one of these familiar markers — things like his standard sign-off "D.B. sealed" or references to known locations like "the cabin at Miller's Pond" — you should assume the key is wrong and go back to phase two. I had a case where the decoding looked perfect for the first 80 percent of the message and then devolved into garbage. The issue was a single-digit error in my date calculation. I had the year right but was off by one day, which threw the entire column permutation out of alignment. Correcting the date fixed it immediately.
Common Pitfalls That Waste Weeks
There are a few recurring mistakes that nearly everyone makes when they first try to work through Badgley's unsolved ciphers. The biggest one is assuming that because a cipher uses a Playfair foundation, the substitution step is independent of the transposition step. It isn't. Badgley sometimes applied the transposition to the digraph pairs themselves before completing the substitution grid. This creates a situation where frequency analysis on the raw ciphertext tells you almost nothing useful because the substitution hasn't been fully applied yet from your perspective. Another mistake is trying to solve these by hand without a systematic tracking method. Badgley's ciphers can run anywhere from 60 to over 200 characters. Doing this manually with no of which key attempts you've already ruled out is a recipe for looping back on yourself repeatedly. I kept a simple spreadsheet with columns for cipher number, assumed date, keyword hypothesis, column permutation result, and plaintext readability score. It sounds tedious but it cut my effective working time roughly in half because I could see at a glance which approaches had already failed.
What Badgley's Work Doesn't Tell You
There's a persistent theory in the small community that follows this material that Badgley encoded something deeper — political secrets, financial information, or references to ongoing criminal activity — in his later ciphers. The evidence for this is thin. The decoded messages mostly deal with personal matters, book recommendations, weather observations, and descriptions of daily routines. The unsolved ciphers haven't yielded anything different yet, but that's partly because they're the hardest ones and we simply haven't spent enough time on them. I'd estimate that maybe 15 to 20 percent of the total material has been thoroughly worked through by anyone. Another thing worth noting: Badgley's methods, while clever for their era, are completely breakable with modern computing. A well-written script can test millions of key combinations in minutes. The reason these ciphers persist in an unsolved state isn't computational difficulty. It's the lack of context. Without knowing the exact date range, the language of composition, or the intended recipient, you're working with more unknowns than the ciphertext itself provides. This is the real bottleneck, and no amount of processing power will solve it.
A Practical Resource List
If you want to actually work through these ciphers yourself, here's what I found useful. The primary source is the 2014 Compendium, which is available through a small print-on-demand service. There's no digital version from the author's estate. The second resource is Badgley's original worksheet notebooks, which were sold at auction in 2019 and are now held in a private collection. Some pages have been photographed and shared on a dedicated forums thread, but the collection is incomplete. For methodology, I relied heavily on Simon Singh's The Code Book for the foundational concepts and then adapted them, plus several academic papers on hybrid cipher systems from the 1970s that Badgley clearly studied. The community around this material is small but active. The main hub is a Discord server and an associated mailing list where people share partial decodings and key hypotheses. It's not a huge community — probably a few hundred active participants at any given time — but the people in it tend to have serious cryptography backgrounds, which keeps the discussion accurate. Joining isn't required, but attempting these ciphers in isolation is significantly harder than working with others who've already hit the same wall.
Bottom Line On Difficulty And Scope
Working through Duff Badgley's cryptography is a legitimate puzzle, not a mystery novel. The ciphers are real, the methods are documented enough to be reverse-engineered, and the solved material proves that the system works as described. The unsolved ciphers remain unsolved primarily because of missing contextual information, not because the cryptographic techniques are beyond current methods. If you approach this with patience, a basic understanding of classical cryptography, and a willingness to write simple automation scripts, you can make real progress. If you're looking for a dramatic reveal or hidden treasure maps, you'll be disappointed. That's not what this is about.
