• tries to define comedy in his book "Laughter", (french Le Rire)
• defines the reasons for laughter:
  • picking up stereotypical human qualities (human),
  • laughter is rather due to intellect rather than emotions (intellectual),
  • laughter is social (social)

One of the problem with philosophers is that the approach is mostly dialectical and it is tough to separate the individual notions aside just by reading or listening to it. Bergson seems to go with a Hegelian-dialectic style, where he tries to reason about comedy in terms of "das Sein" by trying to trace the ties between humans and the environment, where it seems that humor, comedy and laughter does not set humans aside distinctively and that humor would rather treat them as objects. A more comprehensive approach would be a more Kantian-style categorization of humor that sets the reason behind humor aside rather than blending them together using metaphysics as glue.

Associations (thesis, antithesis or synthesis) are too large of a role and Bergson does mention associations, but typically by following humans, or the association of humans, where, in fact, association of two objects with each other, whether in thesis or antithesis, seems more general and well-encompassing of humor in general. For example, a house shaped like a pear is still humorous even if the human role in the association is only… as an observer, to stop there, before going down the Hegelian path.

Here is the beginning of a list, as an example and perhaps antithetical to the Bergson approach, that would name the sources from which humor derives:

• situational humor; compared to other forms mentioned, situational humor is typically non-directed humor where the situation itself is funny, regardless whether humans are involved or not (Bergson involves humans),
• inversions, role-reversals or shuffles ("who's on first?" Abbott and Costello style), whether that involves people, animals or objects; Abbott and Costello is in fact a continuous form of inversion, role-switching and role-reversal where each of the participants (2, in this case) take turns at becoming each-other (which is what we labeled as "shuffle")
• comparisons and contrasts,
• cultural stereotypes,
• hyperbola, belittlings or aggrandizations; general theory adds Nietzschean master-slave dynamics to these labeling them as "superiority", but in fact, most of the time these are humorous due to making appeal to the absurd (perhaps even grotesque) rather than appealing to a certain pre-conceived order (doing so would make them rather fall into the "inversions" category"),
• puns, verbal humor or wordplay, does not seem to involve humans in a metaphysical sense, yet they seem more of a way to bridge the gap between two contexts that are are seemingly disjunct, and the more opposed, the better "technomages" (with "techo", forgetting the real meaning of the word, in terms of mechnization, automation and later computers being distinctly opposed to naturalism and "magic")
• repetitions (perhaps along the lines of puns, the re-application of a rule to a transformed context),
• stating the obvious (perhaps a form of hyperbola and belittlings),
• etc…

In any case the former categorization would suit to deduce the general atomic components but script or a performance (along the lines of stand-up comedy) would involve the combination of all of them. A performance with an audience, requires much more than combining the components, and would even hinge on psychological elements perhaps following Bergson's works closer, in order to gauge what audience is available, what jokes would apply for them and how to lead and/or mislead the audience in order to make them laugh.