MathOverflow is a question and answer site for professional mathematicians. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

In the paper by Masbaum, it was shown that the colored Jones polynomials for a twist knot $K_p$ can be written as

\begin{eqnarray} J_{n}(K_p;q)&=&\sum_{k=0}^{\infty} {\cal C}_{K_p}(k) \frac { \lbrace n-k\rbrace\lbrace n-k+1\rbrace \cdots \lbrace n+k\rbrace} {\lbrace n\rbrace} \cr &=&\sum_{k=0}^{n-1} {\cal C}_{K_p}(k) q^{n k} (q^{-n-1};q^{-1})_k (q^{-n+1};q)_k \ , \end{eqnarray}

where $\def\usc{_}$

\begin{eqnarray} {\cal C}_{K_p}(k) &=&(-1)^{k+1} {q_I}^{k(k+3)/2}\sum_{l=0}^k (-1)^l q^{l(l+1)p} \lbrace 2l+1\rbrace \frac {\lbrace k \rbrace! }{\lbrace k+l+1 \rbrace! \lbrace k-l\rbrace!}\cr &=& (-1)^{k+1} {q}^{k(k+3)/2}\sum_{l=0}^k (-1)^l q^{l(l+1)p+l(l-1)/2}(q^{2l+1}-1) \frac{(q;q)\usc{k}}{(q;q)\usc{k+l+1} (q;q)_{k-l}} \ . \end{eqnarray}

For the trefoil ${\bf 3_1}$, the twist number $p$ is equal to 1. Then the paper says ${\cal C}_{K_1}(k)=(-1)^{k} {q}^{k(k+3)/2}$. My first question: how can it be shown? Namely, does anybody have idea how to prove this? \begin{equation} \sum_{l=0}^k (-1)^l q^{l(l+1)+l(l-1)/2}(q^{2l+1}-1) \frac{(q;q)\usc{k}}{(q;q)\usc{k+l+1} (q;q)_{k-l}}=-1 \end{equation}

In addition, the recent paper (footnote in p.13) showed a simpler form of the colored Jones polynomial for the trefoil \begin{equation} J_n({\bf 3_1};q)=\sum_{k=0}^{n-1} q^{n(k+1)-1} (q^{n-1};q^{-1})_k \ . \end{equation}

My second question: how can one prove that these two expressions are the same? \begin{equation} \sum_{k=0}^{n-1} q^{n(k+1)-1} (q^{n-1};q^{-1})\usc{k} =\sum_{k=0}^{n-1} (-1)^{k} q^{k(k+3)/2} q^{n k} (q^{-n-1};q^{-1})_k (q^{-n+1};q)_k \end{equation}

Although I can check both the identities for $k=2,3$ or $n=2,3$, I have had a hard time to prove them.

Some notations are fixed. \begin{eqnarray} & & \lbrace n\rbrace={q_I}^{n}-{q_I}^{-n}, \; {q_I}^2=q, \\ [n]=\frac {\lbrace n\rbrace}{\lbrace 1\rbrace}, \cr & & \lbrace n\rbrace!=\lbrace n\rbrace\lbrace n-1\rbrace\cdots \lbrace 1\rbrace, \; (x;q)_n=(1-x)(1-x q)\cdots (1-x q^{n-1}) \end{eqnarray}

share|cite|improve this question
up vote 4 down vote accepted

If you look at the paper by Garoufalidis and Le "The colored Jones function is q-holonomic", they prove that the colored Jones function is $q$-holonomic. Using techniques of Zeilberger, one can then verify such identities algorithmically. So you need only check the first few terms are equal, and check that both sides of the equation satisfy the same recursion relation, which is given by the non-commutative $A$-polynomial which is computed for twist knots.

share|cite|improve this answer
Thank you very much for your answer. This is very helpful. – Satoshi Nawata Jul 16 '12 at 2:16

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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