I seem to be having the wrong impression of what $p$ stands for; is $p(x)=x(x+1)(x+2)$ or is it something else? Clarification would be appreciated so that I can complete the lemma below.

Consider the ring $R:=\mathbb{C}[x]/I$ where $I$ is the ideal in $\mathbb C[x]$ generated by $x(x-1)(x-2)$. Show there exists an injective ring homomorphism $R \to \mathbb{C} \times \mathbb{C} \times \mathbb{C}$.

I realize that the first order of business is to show that this is indeed a well-defined ring homomorphism.


For each $\alpha \in \{0,1,2\}$ the evaluation map $\mathbb C[x] \to \mathbb{C}$, $p(x) \mapsto p(\alpha)$, induces a well-defined ring homomorphism $R \to \mathbb{C}$.

The induced map takes $p+I$ to $p(\alpha)$. Thus, for the uniqueness part we only need to check that $I$ is contained in the kernel of the evaluation map of $\mathbb{C}[x]$. That way, if $p=f+r$ for some $r \in I$, so that $p+I=f+I$ in $\mathbb{C}[x]/I$, then $p(\alpha)=f(\alpha)$. It is clear $p(\alpha)=0$ for $\alpha \in \{0,1,2\}$ and so


HINT: Consider the map $$ \Bbb C[X]\longrightarrow\Bbb C\times\Bbb C\times\Bbb C $$ given by $q(X)\mapsto(q(0),q(1),q(2))$. Convince yourself that is a homomorphism, that is surjective, and compute the kernel.

  • $\begingroup$ Please help me understand what $p$ is in this case? Isn't $p(x)=x(x+1)(x+2)$? And so $p(0)=0,p(1)=6,p(2)=24$? $\endgroup$ – abS2tracor Dec 18 '14 at 17:12
  • $\begingroup$ No, just any $p(x)$. The map take any polynomial and assigns to it its values in $0$, $1$ and $2$. $\endgroup$ – Andrea Mori Dec 18 '14 at 17:21
  • $\begingroup$ I have given this a lot of thought and I still don't get it. $\endgroup$ – abS2tracor Dec 18 '14 at 23:47
  • $\begingroup$ @abS2tracor Don't get what specifically? $\endgroup$ – rschwieb Dec 19 '14 at 11:35

This is a special case of the Chinese Remainder Theorem:

If $R$ is a commutative ring with $I, J$ coprime ideals (i.e. $I+J = R$), then $IJ = I \cap J$ and $$\frac{R}{IJ}\cong \frac RI\times \frac RJ$$

via the homomorphism induced by $$R \to R/I \times R/J\\x\mapsto (x+I, x+J)$$

Note that $\mathbb C[X]/(x + \alpha)\cong \mathbb C$ for any $\alpha \in \mathbb C$.


The evaluation map $e_a:p\mapsto p(a)$, where $p\in\mathbb{C}[x]$ and $a\in\mathbb{C}$ is a homomorphism $e_a\colon \mathbb{C}[x]\to\mathbb{C}$. By the properties of products, also $$ f=e_0\times e_1\times e_2\colon\mathbb{C}[x]\to \mathbb{C}\times\mathbb{C}\times\mathbb{C} $$ defined by $$ f(p)=(p(0),p(1),p(2)) $$ is a (ring) homomorphism. You can also check it directly.

What is $\ker f$? It consists of all polynomials $p$ such that $p(0)=0$, $p(1)=0$ and $p(2)=0$. Thus if $p\in\ker f$, we have $$ p(x)=x(x-1)(x-2)q(x) $$ by using the fact that $p(a)=0$ if and only if $p(x)$ is divisible by $x-a$.

Conversely, it is obvious that $x(x-1)(x-2)q(x)\in\ker f$, for all $q\in\mathbb{C}[x]$. Hence $$ \ker f=I $$ where $I$ is the ideal generated by $x(x-1)(x-2)$.

Thus $f$ induces an injective homomorphism $\bar{f}\colon\mathbb{C}[x]/I\to\mathbb{C}\times\mathbb{C}\times\mathbb{C}$.

In addition, we can also note that $f$ (and so also $\bar{f}$) is surjective. Indeed, given any $(a,b,c)\in\mathbb{C}\times\mathbb{C}\times\mathbb{C}$, there exists a polynomial $p(x)$ such that $$ p(0)=a,\quad p(1)=b\quad p(2)=c $$ by Lagrange-Newton interpolation.


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.