Reducing Motifs to Stem Categories

Wherever a motif is listed as, for example, "abduction (any)," this indicates that Tracy's entry for "abduction" includes sub-entries, which have all been combined under their main listing. 

In cases like "abduction" where there are only a few sub-entries, with numerous books associated with each, and potentially interesting distinctions to be discovered, I have created additional columns in the spreadsheet for, for example, "abduction of female" and "abduction of male." The main "abduction" column will then have a "1" entered for each book that contains only a female or only a male abduction, and a "2" entered for each book that contains both. A fourth column also labels each book's abductions as "female," "male," or "both."

However, this fourth column, and the "2" entries, have no bearing on the project at this stage. All graphs are generated based on the "count of" a given motif (i.e., how many rows have an entry of any kind in that column), rather than the "sum of" that motif (which actually adds the numbers that appear in the column), to avoid getting into unanswerable concerns of trying to address how much a given motif appears in each text. (It is possible, after all, that a text might have a large number of abductions, but if they were all of women, it would appear to have fewer than a book with two abductions, one of each gender.) As a result, the sub-categories often add up to more than the total of their stem category: my count of books containing "abduction (any)" is 88, with 69 books containing female abductees and 41 containing male abductees.

In other cases, like blood and poison, the sub-categories were so numerous, with only one or two books included in each, that only the stem category was included.

Creating New Categories

In a small number of cases, I created my own categories; these are always indicated by italics in the spreadsheet. Most, like "mark, identifying (any)" are new stem categories; one is a sub-category which merges multiple pre-existing sub-categories. They are:

• "Mark, identifying (any)," including both "birthmark" and "scar, identifying."
• "Vice (any)," including "drunkenness," "gaming," and "gluttony." Created to look for gendered trends in the relative use of each form of vice; not used in the project at this stage; "drunkenness" and "gaming" considered as their own top-level motifs. (Gluttony excluded for appearing in less than 20 texts.)
• "Fornicator," including "libertine" and "loose woman." Like "vice (any)," not used for this project; "libertine" and "loose woman" treated as their own top-level motifs.
• "Incest mention," including the sub-categories "actual" and "literary flirtation with."
• "Sexually active," as sub-category for "nun," merging "debauched," "eloped," and "ruined," but not "wicked." Not used in the project at this stage.
• "Rape and attempted rape," including "rape" and "attempted rape."
• "Skeletons and skulls," including "skeletons" and "skulls."
• "Suicide mention (any)," including "suicide," and the sub-categories "attempted or contemplated" and "faked or threatened."