To fully comprehend this topic, a basic understanding of multivariable calculus is required. A highly recommended reference is [1], specifically Theorem 2.4.15.
Let
be a simple, undirected graph with vertices labeled as
. Let
be the set of all non-negative real vectors whose components sum to
:

For any vector
, we define the objective function
as the sum of the products of the weights associated with adjacent vertices in
:

Here, each edge
is counted exactly once. The Motzkin-Straus theorem investigates the maximum value of this function over the set
, denoted as:

Remark on Existence: The function
defined in (1) is a polynomial, which implies it is continuous everywhere on
. Furthermore, the domain
is a closed and bounded set (a compact set) in
. According to the Extreme Value Theorem, any continuous real-valued function attains both a global maximum and a global minimum on a closed and bounded domain. Therefore, the maximum value
is strictly guaranteed to exist.
Theorem (Motzkin-Straus, 1965). Let
be the clique number (the order of the largest complete subgraph) of
. Then

Proof. Suppose
is a point in
that maximizes the function

Let
be the set of vertices with positive weights. Suppose
does not induce a complete subgraph (clique). Then, there exist at least two vertices
that are not adjacent in
.
Let
and
be the sum of the weights of the neighbors of
and
in
, respectively. Without loss of generality, assume that
.
We construct a new weight distribution
by shifting the entire weight of
to
:
It is clear that
because the total weight remains
and all weights remain non-negative. The change in the objective function is given by

Since
and
, we have
. However, because
is already a global maximizer, it must hold that
.
The new point
still achieves the maximum value, but the number of vertices with positive weights has decreased by 1 (since
). We repeat this weight-shifting process for any remaining pair of non-adjacent vertices with positive weights. This process must terminate after a finite number of steps. Upon termination, all vertices with positive weights must be mutually adjacent, meaning they form a complete subgraph
.
Let
be the number of vertices in
. Since
is the clique number of
, we must have
. On the complete subgraph
, all vertices are pairwise adjacent, so

Since the sum of the weights on
is
, we obtain

By the Cauchy-Schwarz inequality, we have

This implies 
Since the function
is strictly increasing with respect to
and
, we get:

Now, choose a maximum complete subgraph of
with order
. Assume the vertices of this subgraph are
. By distributing the total weight equally among these
vertices and setting the weights of all other vertices to
, i.e.,

we can compute the value of the objective function at this specific point as:

Combining (3) and (4), we conclude that the maximum value of the objective function on the set
is precisely 
This completes the proof. ∎
Corollary (Turán’s Theorem). Let
be a graph with
vertices and edge set
. If
contains no complete subgraph of order
(meaning the clique number of
does not exceed
), then the number of edges in
satisfies

Proof. Consider the uniform weight distribution
. Clearly,
since all coordinates are non-negative and their sum equals 1.
Evaluating the objective function
at this point, each edge
contributes exactly
. Since the graph
has
edges, we have:

On the other hand, by the Motzkin-Straus Theorem, the value of
at any point in
cannot exceed the global maximum
. Let
be the clique number of
. By hypothesis,
. Therefore:

Combining these two results, we obtain 
The inequality is thus proven. ∎
References
[1] Jerry Shurman, Calculus and Analysis in Euclidean Space (pages 23-56).