# Show that a (1,2)-tensor is a linear function

• KungFu

#### KungFu

Homework Statement
problem:
1a) show that a (1,2)-tensor T is a linear function T: ##E^* x E x E ##
(A (1,2)-tensor is a tensor that takes one vector from E and two covectors form the dual space #E^*#
Relevant Equations
I need some help on where to start.
I know that a tensor can be seen as a linear function.
I know that the tensor product of three spaces can be seen as a multilinear map satisfying distributivity by addition and associativity in multiplication by a scalar.

Your explanation of ##(1,2)## is ambiguous. Is it an element of ##E\otimes E^*\otimes E^*## or an element of ##E^*\otimes E\otimes E##? Let us assume we have a tensor ##T\in E\otimes E^*\otimes E^*##. This means we have
$$T=\sum_{\rho=1}^r U_\rho \otimes v_\rho \otimes w_\rho\; , \; (X,Y) \longmapsto \sum_{\rho=1}^r U_\rho \cdot v_\rho(X) \cdot w_\rho(Y) \in E$$
What kind of function is ##T## then? The situation for ##T'\in E^*\otimes E \otimes E## is accordingly.

Last edited:
Your explanation of ##(1,2)## is ambiguous. Is it an element of ##E\otimes E^*\otimes E^*## or an element of ##E^*\otimes E\otimes E##? Let us assume we have a tensor ##T\in E\otimes E^*\otimes E^*##. This means we have
$$T=\sum_{\rho=1}^r U_\rho \otimes v_\rho \otimes w_\rho\; , \; (X,Y) \longmapsto \sum_{\rho=1}^r U_\rho \otimes v_\rho(X) \otimes w_\rho(Y) \in E$$
What kind of function is ##T## then? The situation for ##T'\in E^*\otimes E \otimes E## is accordingly.

yes, sorry, I had a typo in my description. it is an element of ##E\otimes E^*\otimes E^*##

Can you please explain to me what the capital U and the w and v are in your expression? and also the X and the Y

yes, sorry, I had a typo in my description. it is an element of ##E\otimes E^*\otimes E^*##

Can you please explain to me what the capital U and the w and v are in your expression? and also the X and the Y
The tensor isn't a linear function in that case. What is it?

The tensor isn't a linear function in that case. What is it?

The tensor is a mulitlinear function, but why?, I don't see the arguments. I have been told that it is linear in each variable, due to the distributivity of the field R (real numbers), but can I show it, or is it just by definition ?

You can show it with the definition, see my corrected post #2. I made a copy and paste error and forgot to change ##\otimes## into ##\cdot## which I now did.