All learning is good right?  I need to keep reminding myself of that, because I’m going to have to go back and “re-do” the manifestos I’ve done so far.   My cri de couer yesterday garnered me some expert eyes (thank you Ted Underwoodand Ian Milligan) , which pointed out that Lexomics treated women and women’s as “one.”  My “brain driven” knowledge alerted me to look out for woman v women, but I had not considered the possessive.   I’m going to re-do now using only Voyant.  If you are unfamiliar with Voyant, I urge you to explore it and the online workshops. Created by the folks that brought you TaPor and Hyper Po, it is extremely user friendly.  Eventually I will use Mallet to run the large corpus of manifestos that I’m crowdsourcing here, but for now voyant allows me to play within individual texts and make some nifty visualizations like visual collacator and  collocate clusters even if I’m clueless about the differences between the two! 
So far the most valuable results have been word frequencies, which have generated two hypothesesso far
√ hypothesis one, woman/women most frequent word in things which should count as manifesto
√ hypothesis two,  documents with male/man/men  in top 10 are aimed at criticizing sexism, while those without with male/man/men focus on women’s roles, status.
While I’m super impressed with the word trees created by Aditi Muralidharan, I think the underlying software she uses is far more sophisticated than what I’m getting in many eyes, and she is treeing far larger documents.
So back to square one for me.  

word count 219

Advertisements