Abstract
We address the construction of stable random matrix ensembles as the generalization of the stable random variables (Levy distributions). With a simple method we derive the Cauchy case, which is known to have remarkable properties. These properties allow for such an intuitive method -that relies on taking traces- to hold. Approximate but general results regarding the other distributions are derived as well. Some of the special properties of these ensembles are evidenced by showing partial failure of mean-field approaches. To conclude, we compute the confining potential that gives a Gaussian density of states in the limit of large matrices. The result is an hypergeometric function, in contrast with the simplicity of the Cauchy case.