Is this proof about Jordan algebra's correct?

245 Views Asked by At

Let $A$ be a unital commutative non-associative alternative algebra.

Let $J$ be a unital Jordan algebra. Notice Jordan algebra's are commutative and non-associative by definition.

Conjecture :

The set of $A$ algebra's and of $J$ algebra's are identical.

Is that statement true ?

I propose the following proof :

Let $x,y$ be elements.

Alternative means ( in the context of commutative multiplication )

$$ x(xy) = x^2 y $$

The jordan identity means

$$ x(x^2 y) = x^2 (x y) $$

So we are left with proving those conditions are equivalent.

Start with the first one and transform it into the second :

$$ x(x y) = (x^2) y$$

multiply both sides by $x$ on the left

$$ x^2(x y) = x(x^2 y)$$ done

In the other direction ( changing the second one in the first one )

$$x^2(xy) = x(x^2 y) $$

divide both sides by $x$ on the left

$$ x^{-1} x^2 (xy) = x^{-1} x (x^2 y) $$

simplify

$$ x(xy) = x^2 y $$

QED

Is this proof correct ?

1

There are 1 best solutions below

7
On

You've correctly showed that a commutative alternative algebra is a Jordan algebra. Your proof of the converse is invalid though, and indeed there are Jordan algebras which are not alternative.

Here is on example: the spin factor $J=\mathbb{R}\oplus\mathbb{R}^n$. We think of its elements as formal sums of scalars and vectors from $\mathbb{R}^n$. The multiplication of a scalar with a scalar and of a scalar of a vector are what you think they'd be, and the product of two vectors is their inner product (thus, vector x vector = scalar).

Explicitly, $(a+\mathbf{x})(b+\mathbf{y})=ab+a\mathbf{y}+b\mathbf{x}+\mathbf{xy}=(ab+\langle\mathbf{x},\mathbf{y}\rangle)+(a\mathbf{y}+b\mathbf{x})$.

Exercise. Show the spin factor $J$ is a non-alternative Jordan algebra (for $n>1$).