Let be an odd prime, and consider the quadratic Gauss sum
This sum admits a second description in terms of the Legendre symbol. Each nonzero quadratic residue modulo occurs exactly twice among the numbers
, while each nonresidue occurs zero times. Consequently,
Using the character form we first compute its absolute value:
The terms with vanish. For
, write
. Then
The inner sum is when
, and
otherwise. Since
we obtain
Next we have
\
Therefore and hence
This already determines the magnitude of
Thus the only remaining issue is the sign. The square calculation tells us that there are only two possibilities, but it does not distinguish between them.
Estermann’s idea is to rotate the Gauss sum onto the real axis and then prove that the resulting real number cannot be negative. Define
This factor is chosen so that
Therefore the normalized quantity is real in both congruence classes. Since the square calculation already shows that
has magnitude
, we know that
The problem has now become very concrete: it is enough to prove Any strict lower bound above
rules out the negative possibility and forces
.
Using the symmetry of the terms indexed by and
, we have
Therefore
The expression in the last line has a useful interpretation. The even integers between
and
contribute
The odd integers can be written as , and they contribute
Thus the two factors and
combine into one sum:
This is the form that Estermann estimates. Taking real parts, and using the fact that is real, gives
The constant term causes no difficulty: is either
or
.
The strategy is now visible. For small , the angle
lies in the first quadrant, so the corresponding terms have a definite positive contribution. For larger
, the phases oscillate; one does not try to estimate them term by term, but instead proves that their total contribution is bounded by cancellation.
Set When
, we have
On this interval,
Thus the initial portion of the sum contributes at least
This is the source of positivity in the proof: the first approximately terms all lie in a favorable quadrant and point, in aggregate, in a positive direction.
It remains to control the oscillatory tail
A naive bound by the number of terms would be far too large. Instead, Estermann uses a discrete integration-by-parts argument, which exploits the fact that the quadratic phase rotates more and more rapidly.
Define
The key elementary identity is
Indeed, the difference contains the factor
and was chosen precisely to cancel this sine factor.
Therefore Applying summation by parts gives
This formula has converted the rapidly oscillating sum into a telescoping expression. The complex numbers all have modulus
, while the positive sequence
is decreasing for
. It follows that
Hence
Now . Using the elementary bound
on
, we get
Therefore This is the crucial cancellation estimate. The tail contains almost
terms, but its total size is only of order
. That is exactly the scale needed for the positive initial segment to control the sign.
Combining the initial contribution with the tail estimate, we find
Since ,
Thus
Finally, for every odd prime ,
Consequently,
We have proved But
can only be one of the two values
and
. The negative value is therefore impossible, and so
Undoing the normalization now gives the classical evaluation
The appeal of Estermann’s argument is that it turns a delicate phase question into a robust geometric estimate. The squaring argument reduces the answer to two possible signs. The first summands supply a definite positive bias because their phases lie in the first quadrant. The remaining summands may oscillate in every direction, but summation by parts shows that they cannot collectively cancel enough to overturn that bias. The sign is thus forced by the geometry of the quadratic phases.