It is written in my book that if
T = T_{\nu_1 \cdots v_l}^{\mu_1 \cdots \mu_k} \hat{e}_{(\mu_1)}\otimes \cdots \otimes \hat{e}_{(\mu_k)}\otimes \hat{\theta}^{(\nu_1)}\otimes \cdots \otimes \hat{\theta}^{(\nu_l)} than
T_{\nu_1 \cdots v_l}^{\mu_1 \cdots \mu_k} = T(\hat{\theta}^{(\mu_1)},\ldots, \hat{\theta}^{(\mu_k)},\hat{e}_{(\nu_1)},\ldots ,\hat{e}_{(\nu_l)})
where \hat{e}_{(\mu_{1 \cdots l})} are the basis of a vector space and \hat{\theta}^{(\nu_{1 \cdots k})} are the basis of the correspondent dual vector space.

My question is why does the indexes \mu and \nu switch? In equation 1 \mu is the index of e but in 2 \nu is the index of e. Should't e and \theta be switched in 1?