Prove in Alexandria that primes never end, see repeating remainders in Toulouse, and translate products into series in Saint Petersburg. Continue through zeta zeros in Berlin, average density in Paris and Brussels, and modern sieves and gaps in Oslo, Beijing, Cambridge, and Durham—measuring the distance between ‘a pattern appears’ and ‘it is true infinitely often.’
QUESTION FOR THE ROUTE
Why do primes obey average laws and residue structure even when the next prime is hard to predict, and where must finite computation stop short of proof?
WHAT THIS RIVER DOES NOT CLAIM
The river is not geography. Distance downstream stands for time passing, and the light turns from dawn to dusk as the centuries go by. The objects by each stele are symbols of the kind of event and of how each century band wrote and calculated; they do not reconstruct any real artefact. The land around each stop sketches the natural geography of the scene’s real place, and an iconic building appears only if it already stood in that year. Each scene keeps its real place and evidence basis; open it on the map to read where it happened. This route does not equate primes with a random sequence, a universal key to cryptography, or one person’s discovery. It separates finite samples from infinite theorems, necessary residue conditions from sufficient primality tests, average density from individual locations, and bounded gaps from the twin-prime conjecture. The route line is an edited comparison, not one document’s proven chain of transmission.
WHAT YOU SEE ON THIS RIVER
- Diagram in the sky
- Sieve of Eratosthenes — a number grid with multiples struck out
- Emblem at the source
- A sieve board lit only at the primes
- The real place around each stop
- Around each stele the land takes on the natural geography of that scene’s real place — sea or lake, plain, hills or mountains, the colour of the ground and its common trees — and, where one defines the place, its landform: a volcano, snow peaks, granite domes, a mesa, dunes, a fjord, islands, a rock hill, a gorge or loess terraces. The water near the stop takes the colour of the real river or sea, and the haze the place’s climate. A small globe on the stele marks where it is, with the route from the previous place. Where a city has an iconic building that already stood in the scene’s year, its schematic silhouette rises behind the stop and is named on the card. The land follows today’s terrain and climate as a sketch and the silhouettes are not measured reconstructions. Between stops the river itself stays symbolic.
- A figure board at every stop
- Each board draws the mathematics of that scene. When the boat arrives, the construction is drawn in and the key result rises in red. The drawings are schematic reconstructions, not historical manuscripts.
- Century bands along the banks
- to 499 · Sandstone stele · braziers · earthen villages and beacons · rafts · flocks of birds
- 1450–1749 · Marble stele · iron lanterns · three-arch bridge · windmills and clock-tower towns · sailing ships
- 1750–1899 · Cast-iron plaque · gas lamps · iron truss bridge · factory chimneys, railway and steam train · steamboats
- 1900–1969 · Concrete marker · electric streetlights · concrete bridge · pylons, apartment blocks and radio masts · barges · aircraft
- 1970 onward · Glass marker · LED lights · cable-stayed bridge · glass towers, wind turbines and data centres · ferries · satellites
Where the century band changes, the boat passes under a bridge of the new band. Villages, mills, factories, pylons and towers stand for the technology of each century, not for any real place or architectural style.
01·c. 300 BCE·Alexandria(basis: Composition)
Proving That the Primes Never End — Book IX of the Elements
Proposition 20 of Euclid’s Book IX shows that however many primes are given, those alone cannot account for every number. It is often summarized as ‘multiply them all and add one,’ but the original argument uses a prime divisor of a number one greater than the product. Infinitely many primes is neither a formula for the next prime nor the whole modern theorem of unique factorization.
- Pause and ask
- How can one prove that every finite list of primes must miss another prime?
- How thinking changed
- Move from counting many examples to one general argument that defeats any proposed finite list.
- What we cannot claim
- The exact writing room and date are not known. Proposition 20 does not say that one plus a product of given primes is always prime, and it is not identical to the whole modern unique-factorization theorem.
- This place
- The Alexandrian editorial and teaching tradition associated with the Elements preserved arithmetic propositions as a chain depending on prior definitions and results. (Mediterranean coast · date palms · flat sand · 31.2°N 29.9°E)
- Figure board
- Add a unit DF to ED, the number measured by the given primes A, B, C; a prime G measuring EF is none of A, B, C.
- On the river
- A desk holding a written record · to 499 (Sandstone stele · braziers · earthen villages and beacons · rafts · flocks of birds)
02·c. 240 BCE·Alexandria(basis: Main activity)
Finding by Crossing Out — The Sieve of Eratosthenes
Cross out numbers forced to be composite—multiples of 2, then 3, and so on—and the survivors are prime. The procedure turned a definition into repeatable computation and remains a foundation for finding primes in intervals. Surviving descriptions are later, however, so the exact ancient implementation and date are not reconstructed as modern code.
- Pause and ask
- Instead of testing numbers one by one, can we first remove those forced to be composite?
- How thinking changed
- Turn a static definition into a generative procedure of repeatedly removing multiples—and into a question of computational cost.
- What we cannot claim
- This is not modern pseudocode surviving in Eratosthenes’ own hand. Attribution and approximate date rely on later testimony, and the basic sieve is not the optimal method for every modern primality task.
- This place
- Alexandria’s tables, astronomy, measurement, and collecting institutions supplied audiences and media for organizing long sequences and transmitting procedures. (Mediterranean coast · date palms · flat sand · 31.2°N 29.9°E · landmark: Lighthouse of Alexandria (Pharos) (280 BCE))
- Figure board
- From 1 to 60, crossing out multiples of 2, 3, 5 and 7 in turn leaves only the circled primes.
- On the river
- A place of ongoing work, marked only by the route’s emblem · to 499 (Sandstone stele · braziers · earthen villages and beacons · rafts · flocks of birds)
03·1640 CE·Toulouse(basis: Letter sent)
Seeing Repetition in Remainders Modulo a Prime — Fermat’s Little Theorem
In a letter to Frénicle dated 18 October 1640, Fermat stated that if p is prime and p does not divide a, then a^(p−1) leaves remainder 1 modulo p. No complete proof survives in that letter. The theorem helps test primality, but its converse fails: some composite numbers pass related tests.
- Pause and ask
- Can a repeating remainder pattern modulo a prime be used to test primality?
- How thinking changed
- Focus on recurring remainders rather than quotients, comparing infinitely many integers through finite congruence classes.
- What we cannot claim
- The surviving letter contains no complete proof, and passing a^(p−1)≡1 mod p does not guarantee primality. ‘Little theorem’ does not mean mathematically unimportant.
- This place
- Fermat’s research alongside legal work in Toulouse and a European correspondence network moved unpublished statements and challenges quickly among peers. (Garonne banks · broadleaf trees · flat plain · 43.6°N 1.4°E)
- Figure board
- Powers of 3 divided by 7 leave remainders 3, 2, 6, 4, 5 and return to 1 at the sixth step.
- On the river
- A post holding tied letters · 1450–1749 (Marble stele · iron lanterns · three-arch bridge · windmills and clock-tower towns · sailing ships)
04·1737 CE·Saint Petersburg(basis: Main activity)
Connecting a Product over Primes to an Infinite Series — Euler
Euler developed the divergence of the sum of prime reciprocals and a factorization of a series over all integers into a product indexed by primes. His 1737 work and the Euler product in the 1748 Introductio carried multiplicative structure into analysis. The identity does not directly predict each individual prime.
- Pause and ask
- How can the multiplicative structure of primes appear inside an additive series over all integers?
- How thinking changed
- Leave individual prime tables for a factorization of a series over all integers into an infinite product over primes, making distribution an analytic object.
- What we cannot claim
- The 1737 work and the expression in the 1748 book are not collapsed into one date. Euler products require a convergence-domain distinction and do not list individual primes directly.
- This place
- The St Petersburg Academy’s salary, publication system, and calculating community let Euler move for years among series, products, and integer questions. (Neva delta and gulf · birch and pine · flat · 59.9°N 30.3°E · landmark: Peter and Paul Cathedral (1733))
- Figure board
- The series over all integers is written as a product of one factor per prime; each integer, like 12 = 4·3·1·1, arises once by picking prime powers.
- On the river
- A place of ongoing work, marked only by the route’s emblem · 1450–1749 (Marble stele · iron lanterns · three-arch bridge · windmills and clock-tower towns · sailing ships)
05·1837 CE·Berlin(basis: Publication)
Finding Infinitely Many Primes in Each Eligible Progression — Dirichlet
Dirichlet proved that an arithmetic progression whose first term is coprime to its common difference contains infinitely many primes. The condition matters: not every progression qualifies. Primes were no longer merely scattered points; every admissible residue lane became an infinite distribution problem.
- Pause and ask
- Do primes continue forever inside numbers with a specified remainder?
- How thinking changed
- Refine infinitude of all primes into infinitude within every coprime arithmetic progression, combining characters with analytic tools.
- What we cannot claim
- Coprimality of first term and common difference is essential. The theorem does not locate the next prime in each progression or guarantee short-interval uniformity.
- This place
- Berlin’s university, academy proceedings, and research environment provided a venue to publish and inspect a new combination of number theory and analysis. (Spree banks · pines · flat land · 52.5°N 13.4°E)
- Figure board
- Numbers laid out in four lanes by remainder mod 4: primes keep coming forever only in lanes 1 and 3, the lanes coprime to 4.
- On the river
- A stack of books · 1750–1899 (Cast-iron plaque · gas lamps · iron truss bridge · factory chimneys, railway and steam train · steamboats)
06·1848 CE·Saint Petersburg(basis: Publication)
Bounding Density without an Exact Formula — Chebyshev
Chebyshev rigorously trapped the prime-counting function π(x) on the scale of x/log x and showed that if the limiting ratio existed, it had to equal 1. He did not finish the prime number theorem, but built a strong fence between conjecture and proof.
- Pause and ask
- Without proving an exact asymptotic formula, can prime density still be trapped rigorously on the correct scale?
- How thinking changed
- Replace one plausible approximation with upper and lower bounds and the conditional conclusion ‘if the limit exists,’ raising the level of proof.
- What we cannot claim
- Chebyshev did not prove the full statement π(x)~x/log x. His bounds and conditional result are not retroactively replaced by the later theorem.
- This place
- St Petersburg’s university and academy traditions in probability, number theory, and mechanical calculation supported controlling arithmetic functions through inequalities. (Neva delta and gulf · birch and pine · flat · 59.9°N 30.3°E · landmark: Peter and Paul Cathedral (1733))
- Figure board
- The prime staircase π(x) is fenced between two multiples of x/log x; if the ratio has a limit, it must be 1.
- On the river
- A stack of books · 1750–1899 (Cast-iron plaque · gas lamps · iron truss bridge · factory chimneys, railway and steam train · steamboats)
07·1849 CE·Göttingen(basis: Letter sent)
Recording an Early Observation in a Later Letter — Gauss and the Logarithmic Integral
In an 1849 letter to Encke, Gauss recalled that while studying prime tables in his early years he had come to expect average density 1/log x and regarded the logarithmic integral Li(x) as a better approximation. The familiar claim that he discovered x/log x at exactly age fifteen overstates the precision of a retrospective account.
- Pause and ask
- How did a law like ‘average gap near x is about log x’ emerge from irregular prime tables?
- How thinking changed
- Shift from predicting the next prime to comparing cumulative counts and local average density.
- What we cannot claim
- ‘Discovered at fifteen’ turns an 1849 recollection into an exact birthday event. The empirical approximation is also distinct from the 1896 proof.
- This place
- Göttingen’s observational, computational, and correspondence setting, together with Gauss’s long-kept tables, supported his 1849 account to Encke of an early empirical observation. (Wooded hills · broadleaf trees · small river · 51.5°N 9.9°E · landmark: Göttingen Observatory (1816))
- Figure board
- Prime counts in blocks of 100 slowly thin out; their average density follows 1/log x and accumulates as the logarithmic integral Li(x).
- On the river
- A post holding tied letters · 1750–1899 (Cast-iron plaque · gas lamps · iron truss bridge · factory chimneys, railway and steam train · steamboats)
08·1859 CE·Berlin(basis: Publication)
Hearing Prime Fluctuations through Zeta Zeros — Riemann
Riemann’s Berlin Academy paper extended the zeta function into the complex plane and related its zeros to fluctuations in prime counting. The Riemann hypothesis says that the nontrivial zeros have real part one half. The prime number theorem was proved without the hypothesis in 1896, and even RH would not become a one-line next-prime formula.
- Pause and ask
- How can zeros of a complex function tune the irregular error in prime counting?
- How thinking changed
- Move beyond one average density to analytic continuation, a functional equation, and many oscillations generated by zeta zeros.
- What we cannot claim
- The Riemann hypothesis is not required for the prime number theorem and remains open. Checking finitely many zeros does not prove an infinite claim, and RH would not be a simple formula for individual primes.
- This place
- The Berlin Academy’s monthly reports and election process provided the institutional stage for Riemann’s short, compressed number-theory program. (Spree banks · pines · flat land · 52.5°N 13.4°E)
- Figure board
- Zeta zeros line up on the line of real part one half; adding their oscillations traces the fluctuations of the prime staircase.
- On the river
- A stack of books · 1750–1899 (Cast-iron plaque · gas lamps · iron truss bridge · factory chimneys, railway and steam train · steamboats)
09·1896 CE·Paris(basis: Publication)
Proving the Law of Average Density — Hadamard
Hadamard used complex analysis, including nonvanishing of zeta on the line of real part one, to prove π(x)~x/log x. De la Vallée Poussin found an independent proof in the same year. The symbol ~ says their ratio tends to one; it does not say their finite difference stays small or on one side.
- Pause and ask
- Why does proving that zeta has no zero on a boundary force the average density of primes?
- How thinking changed
- Translate a staircase function on integers into analysis of singularities and zero-free regions in the complex plane.
- What we cannot claim
- Asymptotic equivalence ‘~’ does not mean equality at finite x or a monotonically shrinking difference. De la Vallée Poussin’s independent proof in the same year remains visible.
- This place
- Parisian universities, academies, and journals gathered readers able to publish and compare a long proof linking complex function theory and number theory. (Seine banks · broadleaf trees · flat basin · 48.9°N 2.4°E · landmark: Eiffel Tower (1889), Notre-Dame de Paris (1250))
- Figure board
- Showing that zeta has no zero on the line of real part one forces the ratio of π(x) to x/log x toward 1.
- On the river
- A stack of books · 1750–1899 (Cast-iron plaque · gas lamps · iron truss bridge · factory chimneys, railway and steam train · steamboats)
10·1896 CE·Brussels(basis: Publication)
An Independent Proof Opens the Study of Error — de la Vallée Poussin
In work published through a Brussels learned journal, de la Vallée Poussin independently proved the prime number theorem and used a zero-free region to obtain finer error information. Placing the Paris and Brussels routes side by side reveals the journals, universities, and correspondence that made proofs inspectable—not only the theorem’s name.
- Pause and ask
- Can an independent proof of the same average law also ask a sharper question about its error?
- How thinking changed
- Ask not only whether the theorem is true, but how a zero-free region controls the approximation’s error.
- What we cannot claim
- The pin marks the Brussels publication network and is distinct from the author’s main post at Leuven. The two proofs are not merged into one joint collaboration.
- This place
- Research at Leuven and the Brussels scientific society’s publication network circulated an independent Belgian route distinct from the Paris center. (Beech woods · broadleaf trees · gentle hills · 50.9°N 4.4°E · landmark: St. Michael and St. Gudula Cathedral (1485))
- Figure board
- A thin zero-free region left of the line of real part one, and the funnel it gives that squeezes the relative error of the theorem.
- On the river
- A stack of books · 1750–1899 (Cast-iron plaque · gas lamps · iron truss bridge · factory chimneys, railway and steam train · steamboats)
11·1919 CE·Oslo(basis: Publication)
Showing That the Reciprocal Sum of Twin Primes Converges — Brun’s Sieve
Brun developed a new sieve to prove that the sum of reciprocals of twin-prime pairs converges. Since the reciprocals of all primes diverge, twin primes are dramatically sparser. A finite Brun constant, however, does not decide whether there are finitely or infinitely many twin-prime pairs.
- Pause and ask
- Can one prove how sparse twin primes are without knowing whether infinitely many exist?
- How thinking changed
- Move from a sieve that identifies primes exactly to weighted sieves allowing almost-primes in exchange for upper and lower bounds.
- What we cannot claim
- Finiteness of Brun’s constant proves neither infinitely nor finitely many twin primes. Convergence and number of terms are different questions.
- This place
- The university and learned society in Kristiania, together with contacts made in Göttingen, supported Brun’s combinatorial sieve research and Norwegian publication. (Oslofjord · spruce and birch · wooded hills · 59.9°N 10.8°E · landmark: Akershus Fortress (1300))
- Figure board
- The sum of reciprocals of all primes grows without end, while the sum over twin-prime pairs stays below one constant B.
- On the river
- A stack of books · 1900–1969 (Concrete marker · electric streetlights · concrete bridge · pylons, apartment blocks and radio masts · barges · aircraft)
12·1949 CE·Princeton(basis: Main activity)
Proving the Prime Number Theorem Again without Complex Zeros — Selberg and Erdős
Selberg’s fundamental formula and Erdős’s combinatorial arguments produced an ‘elementary’ proof of the prime number theorem without complex function theory. Elementary describes the permitted tools, not an easy proof. Disputes over priority and communication also make this a shared result rather than a simple lone-victor story.
- Pause and ask
- Does proving the same prime number theorem without complex analysis reveal a different reason it is true?
- How thinking changed
- Replace zeta zeros with symmetric identities for logarithmic prime sums and combinatorial estimates, redrawing the tool boundary called ‘elementary.’
- What we cannot claim
- ‘Elementary’ means neither easy nor historically earlier. The episode is not reduced to a winner’s story in which one person finished everything and the other merely polished it.
- This place
- Visiting research, seminars, and rapid manuscript exchange at Princeton’s Institute for Advanced Study combined Selberg’s and Erdős’s ideas quickly while intensifying priority conflict. (Woods · broadleaf trees · gentle lowland · 40.4°N 74.7°W)
- Figure board
- Without complex zeros: Selberg’s formula weighs prime pairs (p, q) with pq ≤ x by logarithms, counted symmetrically.
- On the river
- A place of ongoing work, marked only by the route’s emblem · 1900–1969 (Concrete marker · electric streetlights · concrete bridge · pylons, apartment blocks and radio masts · barges · aircraft)
13·1973 CE·Beijing(basis: Main activity)
Leaving Only One Extra Prime Factor in Goldbach’s Problem — Chen Jingrun
Chen proved in full detail that every sufficiently large even integer is the sum of a prime and a number with at most two prime factors, extending his 1966 announcement. This is a summit of sieve methods, not the strong Goldbach conjecture that every even integer greater than two is a sum of two primes. ‘Sufficiently large’ and finite exceptions also remain distinct.
- Pause and ask
- If requiring two primes is too rigid, how close can one get by allowing one extra prime factor?
- How thinking changed
- Distinguish primes from almost-primes with at most two factors, replacing an exact conjecture with a quantified near theorem.
- What we cannot claim
- Chen’s theorem says every sufficiently large even N equals p+P₂, where P₂ has at most two prime factors. It is not a proof that every even number is a sum of two primes.
- This place
- The Chinese Academy institute, seminars, and publishing environment in Beijing provided an institutional base for preserving and checking long sieve calculations amid political upheaval. (North China Plain · poplars and pines · flat · 39.9°N 116.4°E)
- Figure board
- A large even N split as a prime p plus P₂, a number with at most two prime factors; the split into two primes stays a question.
- On the river
- A place of ongoing work, marked only by the route’s emblem · 1970 onward (Glass marker · LED lights · cable-stayed bridge · glass towers, wind turbines and data centres · ferries · satellites)
14·2004 CE·Cambridge(basis: Main activity)
Finding Arbitrarily Long Arithmetic Progressions inside the Primes — Green and Tao
Ben Green and Terence Tao proved that the primes contain arithmetic progressions of every finite length. The primes in such a progression need not be consecutive primes. Collaboration across Cambridge and UCLA combined additive-combinatorial structure with pseudorandomness, revealing long order within a sparse set.
- Pause and ask
- Must an increasingly sparse set of primes still contain arithmetic progressions of every desired length?
- How thinking changed
- Do not call primes random; transfer a structure theorem for dense sets into a sparse pseudorandomly weighted setting.
- What we cannot claim
- Primes in the progression need not be consecutive, and the common difference varies by progression. One map pin does not erase Tao at UCLA or the distributed collaboration.
- This place
- Email, preprints, and international seminars connecting Green in Cambridge and Tao at UCLA combined distinct specialties in one proof. (River Cam · willows · flat fen edge · 52.2°N 0.1°E)
- Figure board
- Hidden among the primes are equally spaced runs such as 7, 37, 67, 97, 127, 157 — progressions of any desired length.
- On the river
- A place of ongoing work, marked only by the route’s emblem · 1970 onward (Glass marker · LED lights · cable-stayed bridge · glass towers, wind turbines and data centres · ferries · satellites)
15·2013 CE·Durham, NH(basis: Main activity)
Proving That Primes Come Boundedly Close Infinitely Often — Yitang Zhang
Zhang proved that infinitely many consecutive-prime gaps are smaller than seventy million. Later Maynard–Tao ideas and Polymath collaboration greatly reduced the bound, while the exact gap-two twin-prime conjecture remains open. Finite computation, recurrence below some fixed bound, and infinitely many exact gap-two pairs are three different claims.
- Pause and ask
- Without proving exact gap two, can one show that some fixed bounded prime gap recurs infinitely often?
- How thinking changed
- Reframe the exact gap-two conjecture into a staged breakthrough: first prove that some finite upper bound exists infinitely often.
- What we cannot claim
- Seventy million was an initial upper bound for infinitely many cases, not every consecutive-prime gap. Later smaller bounds still do not prove the exact gap-two twin-prime conjecture.
- This place
- A teaching post, library access, and independent research time at the University of New Hampshire, followed by online Polymath verification, connected a quiet manuscript to rapid collective improvement. (Riverside · pine and broadleaf · gentle hills · 43.1°N 70.9°W)
- Figure board
- Gaps between consecutive primes scatter widely, yet gaps below seventy million recur infinitely often; gap two remains a question.
- On the river
- A place of ongoing work, marked only by the route’s emblem · 1970 onward (Glass marker · LED lights · cable-stayed bridge · glass towers, wind turbines and data centres · ferries · satellites)