Let $X$ be the set of tetrahedron’s edges and $Y$ the set of its faces. To any function $f$ on $X$ with values in a field $K$, we assign a function $g$ on $Y$ defined as

$g(y)$ = $\sum_{x⊂y} f(x)$

That is, the value of $g$ on a face equals the sum of the values of $f$ on the sides of this face.

The above assignment defines a linear map $φ : F(X, K) → F(Y, K)$, where $F(X, K)$ denotes the space of function from $X$ to $K$ and similarly for $F(Y, K)$

Is The kernel of $φ$ has dimension two if $K$ is the field of rational numbers?

Why the dimension of $F(X,K)$ and $F(Y,K)$ are $6$ and $4$ respectively?


2 Answers 2


You can consider the function $1_{l_i}$ that assigns to an edge $x$ the value $0$ if $x\neq l_i$ and $1$ otherwise, for each $i=1, \dots 6$.

If $f$ is a function on $X=\{l_1, \dots , l_6\}$, then you get

$f=\sum_{i=1}^6 f(l_i)\cdot 1_{l_i}$

and so $\{1_{l_1}, \dots , 1_{l_6}\}$ is a base of $F(X, K)$

This is trivial if one observe that $F(X,K)$ corresponds to the free $K-algebra$ generated by the set $X$.

Thus we get the dimension of $F(X,K)$ is equal to $6$, that is exactly the number of tethraedron edges.

An analogous reasoning can be done for $F(Y,K)$, that has dimension $4$ equal to the number of faces.

At this point you can observe $\phi$ is a surjective map. If $g$ is a function on $Y$, then we want to find $f$ such that $\phi(f)=g$, i.e. it is satisfied the following linear system:





and so


that implies, if $K$ has not characteristic $2$, that


If we continue the reasoning, we get


and so





Thus a general solution of the system will be

$f(l_1):=a+ g(F_1)-g(F_2)-\frac{1}{2}k_1$

$f(l_2):= b+\frac{1}{2}k_1$





for some fixed $a,b\in K$.

In other words $\phi$ is surjective and so, by nullity rank theorem, we get

$\dim( \ker(\phi))=6-4=2$


Note that $|X|=6$ and $|Y|=4$ and $F(X,K)=K^6$, while $F(Y,K)=K^4$.

For clarity, let us assume $X=\{(e_1,e_2),(e_1,e_3),(e_1,e_4),(e_2,e_3),(e_2,e_4),(e_3,e_4))\}$ and $Y=\{(e_1,e_2,e_3),(e_1,e_2,e_4),(e_1,e_3,e_4),(e_2,e_3,e_4)\}$ and come to the description of $\varphi$.

It can be described as

$\varphi(\alpha(e_1,e_2,e_3)+\beta(e_1,e_2,e_4)+\gamma(e_1,e_3,e_4)+\delta(e_2,e_3,e_4))\\ =\alpha((e_1,e_2)+(e_2,e_3)+(e_1,e_3))+\beta((e_1,e_2)+\cdots)+\cdots\\ =(\alpha+\beta)(e_1,e_2)+(\alpha+\gamma)(e_1,e_3)+(\beta+\gamma)(e_1,e_4)+(\alpha+\delta)(e_2,e_3)+(\beta+\delta)(e_2,e_4)+(\delta+\gamma)(e_3,e_4)$.

I hope you will be able to conclude from here. Good luck!


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.