Roland Barthes said in an interview in 1973: "As for the sacred outline, I admit having sacrificed at its altars during a certain period in the beginning of semiology. Since then there has been that whole movement challenging the thesis and its format." The dissertation or thesis was, he thought, "oppressive, not to say repressive" just because it consisted of "constraints brought to bear on students by "the myth of the outline" and syllogistic reasoning. These were the reasons why he opted for "an aleatory cutting-up, a découpage (into what I call "miniatures"), or the production of "texts [in the form] of fragments," replacing "logic with chance," albeit in the sense of "controlled accident" as present in "certain Zen-definitions." In other words, he decided that he was "(provisionally) in favor of discontinuous writing" (182).
Since I did not undergo a French university education, I cannot say how justified Barthes' firm coupling of outlines and syllogistic reasoning was subjectively speaking. But even if it were justified biographically or autobiographically for French students of a certain generation to make this connection, there is no necessary relation between these two approaches You can reject Aristotelian logic while holding on to outlines and vice versa. In fact, it appears that at first outlines were developed by thinkers who were questioning the value of syllogistic reasoning (like Ramon Llull). Nor do I believe that outlines and fragmentary writing need be viewed as excluding each other. It is possible to concentrate (provisionally) on short texts and discontinuous writing while still reserving the right to create an outline later. As Barthes notes himself in another interview: "Proust spent half his life producing only fragments, and then, all at once in 1909, he began creating that oceanic flood, Remembrance of things Past" (329). It appears that late in his life he intended to overcome fragmentary writing.
1. Roland Barthes, Against the Grain of the Voice. Interviews 1962-1980. Tr. Linda Coverdale (Berkeley and Los Angeles: Hill and Wang University of California Press), 181.
2. See History of Outlining and Outlines and Meshes. For other posts on Barthes, see here.