Covariance Among latent Variables

[newbie_ari]

In many of the sem examples, we have noticed that covariances are specified between latent variables.

My Questions are:

1) What's the advantage(s) of specifying covariance in sem model? 2) If there's high covariance within 2 variables, does it affect the direct effect? 3) If there's high covariance within 2 variables, does it affect the indirect effect?
4) Lastly, is there any literature specifically explaining the use of covariance in sem models.

Thanks!