Famous Unsolved Puzzles in History
Throughout history, humanity's greatest minds have grappled with puzzles that refused to yield. Some are mathematical conjectures with million-dollar bounties attached. Others are cryptographic manuscripts that have sat undecoded for centuries. A few are ciphers left by murderers, spies, or eccentric artists — taunting solvers with the possibility that the answer lies tantalizingly close.
This guide examines the world's most famous unsolved puzzles: what makes them so difficult, what progress (if any) has been made, and why humanity keeps trying.
1. The Voynich Manuscript (c. 1404–1438)
What Is It?
A 240-page illustrated codex written in an unknown script, discovered in 1912 by Polish antiquarian Wilfrid Voynich. Carbon-dated to the early 15th century, it contains botanical illustrations, astronomical diagrams, and what appear to be medicinal recipes — all encoded in a writing system that has never been matched to any known language.
Why Is It Unsolvable?
The text exhibits statistical regularities that resemble natural language (Zipf's law distribution, consistent "word" structures) yet no decipherment has gained scholarly consensus. Proposed solutions range from an elaborate medieval hoax to a cipher using a constructed language, to a phonemic transcription of an obscure Central Asian dialect.
In 2019, artificial intelligence researcher Gregory Kondrak (University of Alberta) applied machine learning to identify the underlying language as Hebrew — but the resulting "decoded" text was dismissed by most linguists as statistically implausible. A 2022 study in PLOS ONE used network analysis to argue the manuscript contains multiple distinct sections possibly written by different authors, complicating single-cipher theories.
2. The Riemann Hypothesis (1859)
What Is It?
Proposed by German mathematician Bernhard Riemann, the hypothesis states that all non-trivial zeros of the Riemann zeta function — a complex mathematical function linking number theory to complex analysis — lie on a single vertical line in the complex plane (the "critical line" where the real part equals 1/2).
Why Does It Matter?
The distribution of prime numbers is intimately connected to the zeros of the zeta function. A proof of the Riemann Hypothesis would resolve deep questions about how primes are distributed among integers, with cascading implications for cryptography, quantum mechanics, and theoretical physics.
Mathematicians have verified computationally that the first 10 trillion non-trivial zeros all lie on the critical line. Yet no general proof exists. The Clay Mathematics Institute designated it one of its seven Millennium Prize Problems in 2000, offering $1 million to whoever proves or disproves it.
3. The Zodiac Killer's Unsolved Ciphers (1969–1970)
What Are They?
The Zodiac Killer, who murdered at least five people in Northern California, sent four encrypted messages to Bay Area newspapers, taunting investigators with clues to his identity. The first message — a 408-character cipher split across three letters — was decoded by schoolteacher Donald Harden and his wife in just over a week of manual cryptanalysis.
What Remains Unsolved?
The 340-character "Z340" cipher defeated professionals for over 50 years. On 5 December 2020, an international team of amateur cryptographers led by David Oranchak announced a solution using a complex transposition route cipher. But two shorter messages — Z13 (13 characters, possibly the killer's name) and Z32 (32 characters) — remain completely unbroken. Some researchers suspect Z13 may be too short to decrypt without knowing the key system; others believe it intentionally contains no meaningful content.
4. Kryptos — The CIA Sculpture (1990)
What Is It?
Kryptos is a copper sculpture at CIA headquarters in Langley, Virginia, created by artist Jim Sanborn. It bears four encrypted passages totalling 865 characters. The first three passages were decoded between 1999 and 2006 — a mix of Vigenère, Keyed Vigenère, transposition, and what appeared to be a null-key cipher.
What Is K4?
The fourth and final passage (K4), 97 characters long, remains unsolved. Sanborn has provided two tantalising clues: characters 64–69 encode "BERLIN" and characters 70–74 encode "CLOCK." These references to the Berlin Clock suggest a time-based cipher, but no solution has been validated. Sanborn has said he will release another clue when he believes cryptographers are "getting warm."
According to Wired, the sculpture has generated more codebreaking correspondence than any civilian encryption puzzle in history.
5. The Beale Ciphers (1819–1821)
What Are They?
A pamphlet published in 1885 describes Thomas J. Beale, who allegedly buried $43 million worth of gold, silver, and jewels somewhere in Bedford County, Virginia, then entrusted three enciphered papers to an innkeeper. Paper No. 2 — decoded using the United States Declaration of Independence as a key — describes the general location of the treasure. Papers 1 and 3, which supposedly give the precise location and the names of the heirs, remain undecoded.
Hoax or Hidden Fortune?
Many researchers believe the ciphers are an elaborate 19th-century hoax. Cryptanalyst Peter Viemeister argued in 1987 that statistical analysis of Papers 1 and 3 suggests they are random rather than enciphered text. Nonetheless, thousands of treasure hunters have dug up Bedford County farmland in pursuit of Beale's alleged cache.
6. P vs NP — The Computer Science Grand Challenge (1971)
What Is It?
Asked formally by Stephen Cook in 1971, P vs NP asks whether every problem whose solution can be verified quickly by a computer can also be solved quickly by a computer. "Quickly" means in polynomial time relative to input size. If P = NP, thousands of problems currently assumed computationally intractable — including cracking modern encryption — would suddenly be solvable in seconds.
Why Is It Hard?
Most computer scientists believe P ≠ NP, but proving it has resisted every approach for more than 50 years. The difficulty lies in establishing a formal lower bound — proving that no algorithm could possibly solve certain problems efficiently, even ones not yet invented. A 2023 survey of leading theoreticians found 87% believe P ≠ NP, but less than 5% believe a proof will emerge within a decade.
7. The Tamam Shud Mystery (1948)
What Happened?
In December 1948, an unidentified man was found dead on Somerton Beach near Adelaide, Australia. All labels in his clothing had been removed. A hidden pocket in his trousers contained a small piece of paper printed with the Persian phrase "Tamam Shud" ("it is ended"), torn from a rare edition of Omar Khayyam's Rubaiyat.
The Cipher
The book from which the page was torn was found in a nearby car and contained a local woman's phone number — and, on the back cover, five lines of seemingly random capital letters, possibly an enciphered message. Despite cold-case investigations spanning 75 years (including the 2021 identification of the dead man's likely son via DNA genealogy), the cipher has never been definitively decoded, the dead man's identity was only formally confirmed in 2022, and the nature of his death — suicide, murder, or accident — remains unknown.
8. Fermat's Last Theorem's Cousins — The Landau Problems (1912)
While Fermat's Last Theorem was famously proved by Andrew Wiles in 1995, four elementary problems proposed by Edmund Landau at the 1912 International Congress of Mathematicians remain entirely open. They are:
- Goldbach's Conjecture — every even integer greater than 2 is the sum of two primes (unproven since 1742).
- Twin Prime Conjecture — there are infinitely many primes p such that p+2 is also prime.
- Legendre's Conjecture — there is always a prime between n² and (n+1)².
- Near-square primes — are there infinitely many primes of the form n²+1?
These problems are "elementary" in the sense that they require no advanced machinery to state — yet they have resisted the most sophisticated analytic number theory tools developed in 110 years of effort.
Why Some Puzzles Stay Unsolved
Cryptographers and mathematicians identify several recurring reasons:
- Insufficient ciphertext — short messages may be fundamentally undecodable without the key (Z13, Tamam Shud).
- Unknown key material — the best cipher in the world is unbreakable if we don't know what algorithm or key was used (Voynich, Kryptos K4).
- Intentional misdirection — some puzzles may contain deliberate red herrings planted by their creators.
- Mathematical barriers — for open conjectures, we may lack the mathematical tools that will one day make proofs tractable — just as Wiles needed modular forms developed decades after Fermat died.
- Social complexity — when solving a mystery depends on living witnesses or physical evidence, time itself is the enemy (Somerton Man).
How to Get Involved in Active Puzzle Research
Several of these puzzles have active, collaborative research communities:
- The Zodiac Killer Site hosts forums with thousands of citizen analysts.
- The Voynich Manuscript Research community at voynich.nu aggregates scholarly work.
- Kryptos solvers coordinate through dedicated subreddits and the website kryptos-unofficial.com.
- For mathematical puzzles, the Polymath Project at polymathprojects.org coordinates collaborative proofs of open problems in number theory and combinatorics.
Frequently Asked Questions
Has anyone decoded the Voynich Manuscript?
No one has successfully decoded it. Hundreds of cryptographers, linguists, and AI researchers have attempted it since 1912. Some claim it is a hoax; others believe it encodes a real but extinct language or an elaborate cipher system. No proposed solution has gained mainstream scholarly acceptance.
What is the Riemann Hypothesis?
The Riemann Hypothesis proposes that all non-trivial zeros of the Riemann zeta function have a real part equal to exactly 1/2. Unproven since 1859, it is one of the seven Millennium Prize Problems carrying a $1 million reward from the Clay Mathematics Institute.
What is the Zodiac Killer's cipher?
The Zodiac Killer sent four encrypted messages to San Francisco newspapers in 1969–1970. Three remain unsolved. The 340-character cipher (Z340) was cracked in 2020, but the 13-character and 32-character messages are still unbroken.
Is Kryptos fully solved?
No. The CIA headquarters sculpture by Jim Sanborn has four encrypted passages. Three were decoded by 1999–2006. The fourth (K4), containing 97 characters, remains unsolved as of 2026, despite decades of effort by professional cryptanalysts.