The Splintered Mind

Method

* Only authors born 1900 or later are included.

* Each author is only counted once per headline entry (subentries are excluded). In 2010, I found that this generated more plausible results than counting authors multiple times per entry.

* As in 2019, but unlike 2014 and 2010, I include co-authors. Due to the unsystematic formatting of SEP references, this was a somewhat noisy process. To capture last authors, I searched for “and” or “&” in each bibliographic line, if appearing before a “19”, “20”, “forthcoming”, or “in press”, then pulled the text immediately after. To capture second authors that were not last authors, I searched for a second comma before such a date-preceding “and” or “&”, then pulled the text after that. I omitted co-authors in position three or higher unless they were last author. Fortunately for the analysis, co-authorship is relatively uncommon in philosophy compared to the sciences, constituting by my estimate less than 10% of the bibliographic lines.

* Also as in 2019 but unlike 2014, I included editors, but only if their name appears before the date in the bibliographical line. Putting the editor at the front of the bibliographical line highlights the editor’s role or the edited collection as a whole.

* After computerized search and sort, I hand-coded the data, in some cases correcting misspellings and merging authors (e.g., Ruth Barcan = Ruth Marcus), more often separating authors with similar names (e.g., various A. Goldmans and J. Cohens), in a process that involved some guesswork and pattern recognition. Inconsistent syntax and imperfect redundancy removal procedures also created some error, though nothing large or systematic that I noticed. Bear in mind that with about 208,000 bibliographic entries, perfection is not possible! I estimate coding error of up to about +/- 2 entries.