What is the Riemann zeta function?

The Riemann zeta function is one of those mathematical objects whose beginnings give almost no warning of where they are going to lead.

It starts with an infinite addition:

(Choose a value for $s$. Raise every positive integer to that power, take the reciprocal, and add the resulting terms.)

At first sight, there are no obvious prime numbers here. Every positive integer appears. There is no visible geometry, no complex plane, no critical line, and certainly nothing that looks like one of the most famous unsolved problems in mathematics.

Yet all of those things are waiting inside the story.

Long before Bernhard Riemann was born, Leonhard Euler discovered that this same infinite sum could also be built entirely from prime numbers. More than a century later, Riemann allowed its input to move through the complex plane and found that the places where the resulting function becomes zero are intimately connected with the way the primes are distributed.

That second step transformed an already remarkable piece of mathematics into one of the central objects of analytic number theory.

Eventually, one question about those zeros became the Riemann Hypothesis.

But the zeta function deserves to be understood before the hypothesis does. It had a mathematical life before RH, it has a much larger life around it, and the best way to see why its zeros became so important is to begin with the simple infinite sum from which the whole journey grows.

Before Riemann: a famous infinite sum

During the seventeenth and eighteenth centuries, mathematicians became increasingly interested in infinite series: additions with endlessly many terms.

One particular problem acquired a reputation of its own.

What happens if we add the reciprocals of all the square numbers?

(The denominators are $1^2,2^2,3^2,4^2,\ldots$. The terms become smaller and smaller, but there are infinitely many of them.)

The question became known as the Basel problem.

The terms clearly shrink towards zero, so it is reasonable to suspect that the total may settle towards a finite value. But knowing that a sum converges and knowing exactly what it converges to are very different achievements.

The answer gives almost nothing away.

Leonhard Euler found it:

(An infinite sum built only from whole numbers turns out to have an exact value involving $\pi$, the constant more familiar from circles.)

There is already something striking here. Nothing in the list

looks geometric, and yet $\pi$ appears in its exact total.

Euler did not stop with the squares. He considered the wider family in which the exponent itself is allowed to vary:

(Instead of asking only about reciprocals of squares, we introduce a variable $s$ and ask what happens for many different exponents.)

In modern notation, this is written

($\zeta$ is the Greek letter zeta. The symbol $\sum$ means “add all of these terms”; here $n$ runs through the positive integers $1,2,3,\ldots$.)

The Basel problem is now just one value of a much larger object:

What began as a single infinite sum has become a function.

A function is a family of answers

A function is simply a rule that takes an input and returns an output.

For a very simple example, the rule

takes in 3 and returns 9; take in 5 and it returns 25\.

The zeta function is more elaborate, but the principle is the same. The input is $s$, and the output is the value of the corresponding infinite series.

Put in $s=2$:

Put in $s=3$:

Put in $s=4$:

As $s$ increases, the later terms become weaker and weaker. The first few integers dominate, while larger integers contribute less and less.

It is useful to imagine $s$ as a dial.

Turn the dial upwards, and distant terms fade rapidly.

Turn it back towards 1, and they fade more slowly, so more and more of the number line continues to matter.

Eventually we reach

where the series becomes

(This is the harmonic series. Its terms become arbitrarily small, but they do not shrink quickly enough for the total to settle at a finite value.)

That series diverges.

So even at this early stage, the zeta function has a boundary: the simple infinite-sum definition works cleanly only when the real part of $s$ is greater than 1\.

For now, though, we can remain safely in that region.

It is there that Euler finds the first great surprise.

Every integer is hiding inside the primes

Look again at

This appears to be an object built from all the positive integers.

Euler showed that exactly the same function can also be written using only prime numbers:

(The symbol $\prod$ means “multiply”. Instead of adding one contribution for every positive integer, we multiply one factor for every prime.)

This is the Euler product.

It is worth slowing down here.

This identity is not merely a useful alternate formula. It is one of the places where the underlying architecture of arithmetic becomes visible.

Take just the factor belonging to the prime 2:

This factor offers every possible power of 2\.

We may choose no 2 at all.

Or one 2\.

Or two 2s.

Or three.

Now look at the factor for 3:

This does the same thing for powers of 3\.

When we multiply the two expansions together, we make choices.

Take $2^{-s}$ from the first factor and $3^{-s}$ from the second:

(Choosing one factor of 2 and one factor of 3 constructs the integer $6=2\times3$.)

Choose $2^{-2s}$ and $3^{-s}$:

Choose $2^{-3s}$ and $3^{-2s}$:

Now add the factor for 5, and we may include any power of 5\.

Then 7\.

Then 11\.

Then every prime.

Eventually the product generates every positive integer.

And it does so exactly once.

The reason is the fundamental theorem of arithmetic: every integer greater than 1 has a unique prime factorisation, apart from the order in which the factors are written.

For example,

(To construct 360 inside the Euler product, choose the $2^3$ term from the 2-factor, the $3^2$ term from the 3-factor, the $5$ term from the 5-factor, and 1 from every other prime factor.)

There is only one prime-factor recipe for 360\.

There is only one for 84\.

There is only one for 17\.

There is only one for every positive integer greater than 1\.

So when the Euler product is multiplied out, the original zeta series reappears.

The sum says:

“Take all the integers.”

The product says:

“All the integers can be rebuilt from the primes.”

And the zeta function is the place where those statements become the same mathematics.

(This is why the connection between zeta and the primes is structural, not merely numerical. Unique prime factorisation is built directly into the function.)

It is hard to overstate how important this is.

The prime numbers appear irregularly when we look at them one by one:

Their gaps vary. Their individual positions are difficult to predict.

Yet when we package all the integers into the zeta function, the primes emerge not as a messy subset of them, but as the irreducible factors from which the entire object can be reconstructed.

Euler had exposed a deep bridge between multiplication and analysis.

Riemann would walk much further across it.

Riemann changes the question

So far, we have allowed $s$ to be an ordinary real number: 2, 3, 4, perhaps 1.5.

Bernhard Riemann asked what happens when $s$ is allowed to be a complex number.

That sentence is easy to read and surprisingly easy to skate past.

But this is the moment where the character of the zeta function changes, so it is worth taking our time.

We need to understand what a complex input actually does.

Start from scratch: numbers that can point

An ordinary real number lives on a line.

Positive numbers extend in one direction from zero, negative numbers in the other.

A complex number adds a second direction.

It is usually written

where $i$ satisfies

Instead of picturing $a+bi$ as a point on a line, we can picture it as a point on a plane.

The real part, $a$, tells us how far to move left or right.

The imaginary part, $b$, tells us how far to move up or down.

So

can be pictured as the point three units to the right and two units up.

Now draw an arrow from the origin to that point.

That arrow has a length and a direction.

This is the first important shift.

A positive real number tells us essentially one thing: size.

A complex number can naturally carry two related pieces of information: how large something is, and which way it points around a plane.

(When mathematicians speak of the “phase” of a complex number, they are describing this directional information — its angle around the complex plane.)

That gives us a picture we can work with.

Now we can begin to crossfade back into the zeta function.

A ring at the centre of a table

Imagine a perfectly smooth, frictionless table.

At its centre sits a tiny metal ring, free to move in any direction.

Many extremely light strings are attached to the ring, and each string runs away across the table to someone holding its far end.

Each person can pull.

If every person pulls directly to the right, then only the strengths matter.

Suppose the pulls have strengths

The total pull to the right is simply

We have represented $\zeta(2)$ physically.

Nothing mathematically new has happened yet. A positive number is just being drawn as an arrow pointing right.

But the picture now gives us another possibility.

We can turn the strings.

Suppose one person pulls to the right, another upwards, another diagonally downwards, another partly to the left.

Now the total effect depends on more than the strength of each pull.

Direction matters too.

Two pulls can reinforce each other.

Two can partly oppose each other.

Several smaller pulls can combine against one larger pull.

The ring moves according to the vector sum of all the forces.

(Complex-number addition behaves in exactly this way: arrows combine according to both their lengths and their directions.)

This is not yet the zeta function.

It is the physical structure we need in order to understand what happens when the zeta function receives a complex input.

First: what controls the strength?

Write the complex input as

($\sigma$ — sigma — is the real part of $s$. The quantity $t$ controls its imaginary part.)

For a moment, set

Then $s=\sigma$, and the $n$th term in the zeta series is simply

If $\sigma=2$, the strengths are

If $\sigma=3$, they are

So the real part $\sigma$ determines how quickly the contributions weaken as $n$ becomes larger.

In our ring-and-strings picture:

$\sigma$ controls the length of each pull.

That is one half of the complex zeta term.

The other half controls direction.

How do we write a rotation mathematically?

Before returning to zeta, consider a single arrow of length 1\.

Point it directly to the right.

Now rotate it by some angle $\theta$.

Its horizontal component is

and its vertical component is

So the corresponding complex number is

Euler’s formula tells us that this is exactly the same as

($e^{i\theta}$ is therefore a compact way of writing a unit-length arrow pointing at angle $\theta$.)

This formula is one of the great bridges inside complex mathematics.

The exponential function, trigonometry and rotation all meet in a single expression.

So if we want to describe a pull of length $R$ pointing at angle $\theta$, we can write

($R$ controls the length. The exponential factor controls the direction.)

Now we have precisely the language needed for a complex zeta term.

Let the zeta pulls rotate

Return to

The $n$th contribution to the zeta series is

Substituting $s=\sigma+it$ gives

Using the ordinary rules of exponents,

And because powers can be written using exponentials and logarithms,

So altogether,

(The zeta term has now separated into two clear jobs: $n^{-\sigma}$ determines its length, while $e^{-it\log n}$ determines its direction.)

This is the point where the physical picture and the mathematics meet.

The quantity

gives the strength of the $n$th pull.

The quantity

is a unit-length arrow pointing at angle

So the complete contribution

is a pull whose strength is $n^{-\sigma}$ and whose direction is $-t\log n$.

Nothing has been smuggled into the analogy.

That is what the equation itself says.

What happens when we turn the $t$ dial?

Now hold $\sigma$ fixed and imagine changing $t$.

At

every angle is zero, so every contribution points to the right.

We are back to the ordinary positive series.

But as $t$ changes, the arrows rotate.

Crucially, they do not all rotate together.

The angle of the $n$th term is

Since

are all different, the arrows turn at different rates.

The contribution associated with 2 follows one angular rhythm.

The one associated with 3 follows another.

The one associated with 10 another again.

As $t$ moves, the entire arrangement continually changes.

Some arrows become more closely aligned and reinforce each other.

Others begin to oppose one another.

A group that pointed mostly one way at one value of $t$ may rearrange into something very different at another.

Where the original series converges, the zeta function is the total of all these contributions:

(The zeta value is the resulting complex arrow obtained by adding infinitely many individual arrows with mathematically determined lengths and directions.)

This gives us a much richer picture of what a complex-valued function is doing.

Its value can point in any direction in the complex plane.

It can have a large magnitude or a small one.

Its real and imaginary parts can reinforce or oppose each other in intricate ways.

And if the total arrow ever has exactly zero length, then the function is zero.

(A complex zero requires both the horizontal and vertical components to vanish together. In the vector picture, the complete collection of contributions has no net resultant.)

That sounds tantalisingly close to the famous zeros of zeta.

But this is exactly where we need to stop.

Where the picture stops

The direct zeta series

converges only when

In our notation, that means

And in that same region, Euler’s product tells us something decisive:

None of those factors is zero.

So zeta itself has no zeros there.

The famous non-trivial zeros do not occur in the region where our original “add all the pulls” series converges.

They lie farther left, in the critical strip:

(The critical strip sits between the vertical lines with real coordinates 0 and 1.)

So we must not picture a non-trivial zero as though we simply took the original series into the critical strip and watched its strings balance perfectly.

The original series does not converge there.

The ring-and-pulls model has done useful work: it has taught us how a complex zeta term acquires magnitude and direction, how phase enters the function, and how complex cancellation works.

But it has reached its boundary.

To move farther into Riemann’s landscape, we need a new idea.

A formula is not always the whole object

Suppose someone gives you a map of southern England.

The map may stop at a particular edge, but the country does not stop there.

The limit belongs to the map, not necessarily to the landscape.

That analogy is imperfect — mathematics is far stricter than cartography — but it captures the conceptual move we need.

The infinite series is one way of describing the zeta function.

It may cease to work before the underlying analytic object ceases to exist.

To see how this can happen mathematically, consider a simpler series:

For

the powers shrink, and the series converges to

Take

Then

while

Both descriptions agree perfectly.

Now try

The original series becomes

which plainly diverges.

But the expression

still makes sense at $x=2$:

This does not mean that

is an ordinary convergent sum whose value is $-1$.

That would be the wrong conclusion.

What it tells us is that the infinite series was one representation of a function, valid only in a certain region, while another representation of that same function continues beyond it.

(The formula has reached its boundary; the function has not necessarily reached its own.)

This is the intuition behind analytic continuation.

But here the map analogy must be left behind, because analytic continuation is not merely a matter of drawing more territory by hand.

Analytic functions are extraordinarily rigid.

If an analytic continuation exists, the values already known in the original region determine the continuation uniquely.

(We are not free to invent whatever values we like outside the original domain. The existing analytic structure constrains the extension.)

This is what allows the Riemann zeta function to escape the boundary of its defining series.

It begins with

but analytic continuation extends $\zeta(s)$ to almost the entire complex plane.

There is one exceptional point:

At that point, zeta has a simple pole.

(A pole is a controlled kind of singularity: rather than taking an ordinary finite value, the function grows without bound in a precisely structured way.)

So when mathematicians evaluate $\zeta(s)$ inside the critical strip, they are not pretending that the original divergent series has begun converging again.

They are working with the unique analytic continuation of the function that the series originally defined.

That distinction is fundamental.

What Riemann actually added

Euler had already uncovered the reciprocal-power series and its product over the primes.

Riemann’s achievement was not to conjure the zeta function from nothing, but to recognise what became possible once it was treated as a function of a complex variable.

In his 1859 paper On the Number of Primes Less Than a Given Quantity, Riemann studied the continued zeta function in the complex plane and connected its analytic behaviour, especially its zeros, with formulas describing the distribution of primes.

(Riemann’s paper was about prime numbers. The zeros mattered because they entered the machinery used to understand how primes are distributed.)

This is one reason the name Riemann zeta function is appropriate even though crucial parts of its history begin with Euler.

Euler revealed the arithmetic structure.

Riemann opened the complex landscape.

And in that landscape, a new symmetry appeared.

The symmetry around one-half

The analytically continued zeta function satisfies a functional equation relating its value at $s$ to its value at $1-s$.

One common form is

(The precise factors are less important here than the structural message: values at $s$ are tied to values at the reflected point $1-s$.)

The reflection

sends a real part $\sigma$ to

So a point with real coordinate $0.2$ is paired with one at $0.8$.

A point at $0.3$ is paired with one at $0.7$.

And the point that does not move under this reflection is

That makes the vertical line

the natural centre of the symmetry.

(The critical line is not an arbitrary line chosen because one-half looks aesthetically pleasing. It sits at the fixed centre of the transformation $s\mapsto1-s$.)

The same symmetry can be packaged even more cleanly by combining zeta with additional factors into the completed xi function.

But for our present journey, the essential fact is already visible:

the complex geometry of zeta has a distinguished middle line.

Now we can meet the zeros.

What is a zero?

A zero of a function is simply an input at which the output is zero.

For example,

has a zero at

because

The same idea applies to zeta.

We ask for values of $s$ such that

Some are comparatively straightforward.

They occur at

and are called the trivial zeros.

The name does not mean that they are uninteresting. It means that their existence follows relatively directly from the functional equation.

The others are the non-trivial zeros.

They are known to lie inside the critical strip:

And now the Riemann Hypothesis can finally be stated:

for every non-trivial zero.

(RH says that every non-trivial zero lies exactly on the symmetry line through the middle of the critical strip.)

That is the conjecture.

But having approached it through the zeta function rather than meeting it in isolation, the statement looks different.

We know where the function came from.

We know why primes are built into it.

We know what complex inputs add.

We know why the original infinite sum is not enough.

We know why the critical strip requires analytic continuation.

And we know why one-half is structurally distinguished.

There is one major bridge left to cross.

Why should the positions of the zeros tell us anything about the primes?

The journey turns back towards the primes

Euler’s product already tells us that zeta contains the primes.

Riemann discovered something deeper: the zeros of the analytically continued function appear in formulas that describe how the primes are distributed.

There are several versions of these so-called explicit formulas, but their broad structure can be understood without writing the most technical form.

Very roughly:

prime-counting behaviour \= smooth average trend \+ corrections associated with the zeta zeros.

(The primes become less common according to a broad statistical law, while the zeros help describe the fluctuations around that smooth trend.)

This is the point at which the journey closes its loop.

We started with all the integers.

Euler showed that the primes were hidden inside the same function.

Riemann moved that function into the complex plane.

Its zeros then turned out to carry information back to the distribution of the primes.

To see why the position of a zero matters, write one as

($\beta$ is the zero’s horizontal coordinate in the complex plane; $\gamma$ is its vertical coordinate.)

Terms associated with such a zero in explicit formulas involve expressions related to

Now use the same mathematical structure we encountered earlier:

We have seen this shape before.

Earlier,

There, one factor controlled size while the other controlled phase.

Exactly the same separation appears here.

In

the factor

controls scale, while

produces oscillation.

(The imaginary part of a zero contributes oscillatory behaviour. The real part affects how large that contribution can become as $x$ grows.)

This is why the horizontal position of a zero matters.

It is not merely a coordinate on a diagram.

It controls the scale of a contribution to the fluctuations in prime-counting formulas.

Now the Riemann Hypothesis acquires a deeper meaning.

If every non-trivial zero satisfies

then the zero-driven oscillations are all constrained by that same critical horizontal scale.

If a zero existed farther to the right, with a larger real part, its corresponding contribution could support larger fluctuations.

So RH is often described, very roughly, as placing a strong restriction on the irregularity of the primes.

It does not make the primes evenly spaced.

It does not turn them into a repeating pattern.

It tells us something subtler: how far the actual distribution of primes is allowed to wander from its smooth average behaviour.

(The Riemann Hypothesis is not a claim that the primes are regular. It is a claim about the size of their irregularity.)

This is the deepest bridge in the story.

The primes shape the zeta function through Euler’s product.

The zeros of the zeta function, in turn, shape our understanding of the primes.

The whole function in compact form

For readers who want the mathematical skeleton without the slower route, the central facts can now be collected quite briefly.

For

the Riemann zeta function is defined by

In the same region,

The first expression is a sum over every positive integer.

The second is a product over every prime.

The function extends meromorphically to the entire complex plane, apart from a simple pole at

Its functional equation links $s$ with $1-s$ and gives the critical line

a natural central role.

Its trivial zeros occur at the negative even integers.

Its non-trivial zeros lie in

And the Riemann Hypothesis asserts that every one of those non-trivial zeros lies on

That is the formal structure beneath the longer story.

The zeta function did not stop with Riemann

The fame of the Riemann Hypothesis can make the zeta function look as though its principal purpose is simply to house one unsolved conjecture.

Its mathematical life is much broader.

One immediate continuation came in 1896, when Jacques Hadamard and Charles Jean de la Vallée Poussin independently proved the prime number theorem using complex-analytic arguments in which zeta played a central role.

One standard way to express the theorem is

($\pi(x)$ counts the number of primes up to $x$. The symbol $\sim$ means that $x/\log x$ gives an increasingly accurate description of its overall scale as $x$ becomes large.)

Their proofs depended crucially on showing that zeta has no zeros on the line

So long before RH was solved — and it remains unsolved — information about the location of zeta zeros was already powerful enough to settle a fundamental theorem about prime distribution.

The zeta function also continued to generate remarkable questions through its individual values.

Euler found exact formulas for every positive even integer:

and so on.

The odd values behave much more mysteriously.

For example,

is known as Apéry’s constant, after Roger Apéry proved that it is irrational.

No comparably simple formula in powers of $\pi$ is known for $\zeta(3)$.

Even the individual outputs of the function continue to hold unanswered questions.

How far the idea travelled

Perhaps the strongest sign of the zeta function’s importance is that mathematicians did not merely continue studying $\zeta(s)$ itself.

They generalised the architecture behind it.

Dirichlet $L$-functions adapt similar ideas to questions about primes in arithmetic progressions.

Dedekind zeta functions carry information about algebraic number fields.

The Hurwitz zeta function shifts the underlying series.

Much larger families of $L$-functions now sit at the heart of modern number theory.

(The Riemann zeta function became a prototype: arithmetic information is packaged into an analytic object, and the object’s products, poles, symmetries and zeros are then studied.)

The same broad machinery reaches beyond number theory.

Zeta functions appear in spectral geometry and mathematical physics, where they can package information about eigenvalues — the characteristic frequencies or energy levels associated with mathematical operators.

Related zeta-function methods also appear in regularisation, where an analytically continued function is used to associate a controlled value with an expression that does not converge in the ordinary sense.

(Regularisation does not mean that a divergent series has secretly become convergent. It defines a related quantity through analytic continuation, and the distinction matters.)

Such techniques appear in areas of quantum theory and in calculations associated with the Casimir effect.

There is also a striking connection between the statistical behaviour of the non-trivial zeta zeros and ideas from random matrix theory and quantum chaos. Those parallels have inspired deep research into whether the zeros might someday admit a spectral interpretation.

One famous possibility, associated with the Hilbert–Pólya idea, is that the relevant zero data could arise as the spectrum of a suitable self-adjoint operator.

(A successful spectral construction of the right kind could explain why the relevant quantities are real and might therefore force the zeros onto the critical line. No such complete construction is currently known.)

These later connections are important, but they should not obscure the original thread.

The zeta function became influential because it discovered a way of translating between different mathematical languages.

Integers became analysis.

Primes became a product.

Complex geometry became information about arithmetic.

That basic pattern proved fertile far beyond the question from which it began.

Why this function became so important

Seen from a distance, the journey is remarkable.

A question about an infinite sum leads to a family of reciprocal powers.

Unique prime factorisation turns that family into an infinite product over primes.

Riemann moves the function into the complex plane.

Analytic continuation opens territory the original series cannot reach.

The functional equation reveals a symmetry.

The zeros enter formulas for prime distribution.

And one conjecture about their location becomes the Riemann Hypothesis.

What matters is not merely that many interesting facts happen to cluster around the same formula.

The same structure keeps reappearing from different directions.

The positive integers give us the series.

Their prime factorisation gives us the Euler product.

The complex continuation gives us the zeros.

The zeros lead us back to the primes.

That closed loop is why the zeta function feels less like an isolated formula and more like a meeting place.

It links arithmetic and analysis so tightly that information can travel from one side to the other.

Why the zeta function comes before the Riemann Hypothesis

The Riemann Hypothesis is often introduced first, with the zeta function appearing as technical machinery needed to state it.

Conceptually, the order is more illuminating the other way around.

The zeta function came first.

Its relationship with the primes came first.

Riemann’s complex extension came next.

Then came the zeros.

Only then did the conjecture about their location appear.

Seen in this order, RH stops looking like an arbitrary claim about mysterious points on a diagram.

The line

is already the centre of a symmetry.

The zeros already influence prime-counting fluctuations.

And the function whose zeros we are studying is already built, through Euler’s product, from the primes themselves.

The Riemann Hypothesis asks whether this entire structure is more rigid than we currently know how to prove.

Where Riemann Console enters the story

Riemann Console begins from this same classical mathematical landscape.

The public research programme explores representations and structures related to the Riemann zeta function, its completed xi form, prime-number information, transforms, positivity questions and other analytic constructions.

Those explorations sit downstream of the classical zeta function. They do not replace it, redefine it or change the status of the Riemann Hypothesis merely because an interesting representation has been found.

Some research routes survive testing.

Some fail.

Some produce bounded results.

Some encounter counterexamples and have to be retired.

The zeta function remains the common mathematical ground beneath them.

For a concise definition, the Riemann zeta function glossary entry remains the quickest reference.

For the conjecture about the location of its non-trivial zeros, continue with What is the Riemann Hypothesis?.

And having followed the zeta function all the way from a simple infinite sum to the edge of RH, we are ready for the next question.

Euler showed that the primes are built into the zeta function.

Riemann showed that its zeros can describe their fluctuations.

How can the zeros of a complex function possibly tell us where the prime numbers are?