I am planning in running a Ph.D. student seminar next year on representation theory in the spirit of [MIT Kan's Seminar][1] where students give lectures on classical articles on representation theory that are typically not covered in a first course but that can none-the-less be covered in one week or so. I would like to amass a list of 12-20 articles or book chapters that should compose such a seminar. Examples would be 

*Demazure Inventiones 33 (1976) 271-272* "A very simple proof of Bott's theorem", 

*A. Beilinson, J. Bernstein, Localization de g-modules, C.R. Acad. Sci. Paris, 292 (1981), 15-18.*

*Chapter 4 of Chriss-Ginzburg "Representation Theory and Complex Geometry".*

and **non-examples** should be the geometric Satake isomorphism or Zhu's modularity of characters of vertex algebras. 

I am not sure if this question goes here or not, but ME seemed like the wrong place to ask. Since I found similar questions here like http://mathoverflow.net/questions/101511/a-request-for-suggestions-of-advanced-topics-in-representation-theory and http://mathoverflow.net/questions/2755/a-learning-roadmap-for-representation-theory I figured it was alright. 

**Edit:** as Tobias and Jim asked for background, I would want this class to be a second year graduate class in representation theory for students with different backgrounds, however, in general I would require students to be familiar with the standard first year classes like finite dimensional Lie algebras over $\mathbb{C}$ (say Jim's GTM book), finite dimensional compact groups (as in the first chapters of Knapp's Lie groups beyond an introduction), the basics of algebraic geometry as in the first 3 chapters of Hartshorne. And hopefully some knowledge of differentiable manifolds as in Warner and/or complex geometry as in Wells. 

The focus of the seminar may/can vary from year to year, as does Kan's seminar, I myself would rather be more about geometric representations as the list supplied suggest. Finally the definition of *classical* is left intentionally vague as I would want it to be "those articles that most in the field have read or should have read". 

  [1]: http://ocw.mit.edu/courses/mathematics/18-915-graduate-topology-seminar-kan-seminar-fall-2014/