Today’s manifesto, The 1977 Combahee River Collectivestatement

Update, came back to work on CRC since manifesto is extremely important as one of the earliest and most cogent statements of black feminism.

The collocates off black are women, white, feminism, feminists

Collocate cluster

This visualization may be one of the most useful I’ve generated.  The clusters replicate almost perfectly the threads of the document.

I alos love the way the collocate

In terms of my hypotheses generated so far

 hypothesis one, woman most frequent word, documents that function as manifestos for feminism will have women/woman very high or at top of frequent word YES
√ hypothesis two,  presence of male/man/men high in frequency indicates document that will be attacking men’s role in oppressing women  ??
√ hypothesis three  woman/women/female  indicates document will focus on women’s status or what women should do YES
√ hypothesis four documents labeled as manifesto will “look” different digitally than those not labeled as manifestos ??

 Technical question, why did I get two word frequency results?

First I tried simply uploading the text to voyant and used its stop word removal function before generating word frequency.  Then I ran text through wheaton removing stop words and punctuation then moved to voyant to generate  word frequency chart

Voyant only Wheaton then Voyant
black 88   black 88
women 44   women 61
political 25   political
oppression 20   oppression 20
politics 20   politics 20
white 19    white 20
work 18    feminists 17
feminists 17    feminist 16
men 17    lives 14

For the sake of consistency I’m going to work with the text that went through Wheaton Lexomic’s scrubber first as that is what I’ve done for all the other manifestos.

Moving that text into many eyes reveals the following word trees

Not surprisingly black most frequently associated with women followed by feminist

women leads to movement and other chains that relate to organizing

word tree “black women”

as I work with word trees, I’m fascinated by them, but increasingly worried about my “on the fly” interpretation of them.  So far all I can find are CS articles that confuse the hell out of me.  Has any humanities person written about interpreting them?  

@miriamkp has some intersting things to say about interpreting topic modeling here in which she argues

By now it should be abundantly clear that no part of this process is “scientific”; it’s just one way of getting your head around a large body of text. So there’s no right or wrong topic name, just schemas that do and don’t help you find interesting features of the text you’re looking at