Visualizing Gender in HIstory of Woman Suffrage

nb “Under this name she is fitly described”: A Digital History of Gender in the History of Woman Suffrage, Women and Social Movements, forthcoming March 2015

Moving towards publication, jump to next red if you're curious where I started. 

project inspired by this quote and conducted with the kind assistance along the way of Ellen DuBois

Subjecting the subject of woman suffrage to a gender history approach must surmount the obstacle that the term “gender” is absent from the primary sources. But this does not mean that the underlying concepts packed into the contemporary use of the term are absent. Tani Barlow has approached the history of women’s rights (including suffrage) in China by tracing the shifting Chinese language terms for woman/female. The English language does not provide such an easy entry into this project – though the terms for the political movement were numerous, shifting, and subtly different in connotation – woman suffrage, equal suffrage, votes for women, home protection, women’s suffrage, etc. Ellen DuBois  

More gender in History of Woman Suffrage June 11

Female appears in about 20% of the male authored items in the History of Woman Suffrage while male appears in about half of the male authored files. I first looked at significant patterns of usage, clusters (table 2). Male clusters reflect a legal discourse centering on men’s role as “citizens” and the rights thereof “male voter. ” “Female suffrage” is one ways the suffrage demand was articulated (graph 3). Other clusters pertain to apsects of the movement as well “female delegates” to the conventions for example. Interestingly two "sutdents" and "physicians" refer to suffrage movement overlaps with women’s education.

Table 2

male sex	female sex
male and	female suffrage
male citizens	female delegates
male citizen	female students
male members	female physician
male voters	Female portion

I got curious about who was using female and male in the corpus, so before I poked around the above results anymore I ran that analysis.  When I visualized those results by year I noticed female peaked 1850s (data viz 3,  male1880s (data viz 4). Next up, digging to figure out why!

DataViz 3

Data Viz 4

While in my iterative process of moving from the computer analysis of the texts, to reading the files, I found this quote I LOVE that speaks gender

“Man-masculine is not endowed by his Creator with certain inalienable rights because he is male, but because he is human; and when, in virtue of our strong andsuperior physical capacity, we deny to man-feminine the rights which are ours only in virtue of our humanity" C.L. Sholes Minority Report, Committee on Expiration and Reenactment of Laws, Wisconsin, 1853

Today I spent some time with male and female in context which caused me to refine my preliminary diarchonic analysis of Data Viz 3 and 4.    Uses of male divide pretty sharply at the Civil War, due to the debates over first the Reconstruction Era amendments to the Constitution and then to various State battles to win women's suffrage, which hinged on the word male.  The use of male prior to the Civil war in the male corpus is almost exclusively religious.  The uses of female don't exhibit these stark divides, although I found it fascinating that prior to the Civil War, neither 

female citizen nor female suffrage appeared in the male corpus.  The most predominate cluster pre-Civil War was female sex or just plain female. 

These distinctions are far less clear in the female corpus, which is smaller in usage of female  and male (see keyness table below).    Is is relative absence of female an effort by women authors to draw attention away from their sex?  Is the relative absence of the religious connotations seen in the male corpus a reflection of the fact that women were excluded from the clergy in virtually all religions during this time and not seen as fit to offer religious pronouncements?

Leaving behind why the women don't use words like men, I started to explore the female corpus.  I looked at a distinction I had previous identified (table 1) in the female corpus,  her husband (graph 2).   I examined lines not with only her husband but collocates, which is to say her within 5 words to the right or left of husband.   As usual Heather Froelich saved my ass here by pointing out I might want to reconsider methodology.  See her excellent piece paper on gender-specific possession.

her husband is spread over a long range of dates, from 1848-1902, and appears in about 13% of items.  her husband appears disproportionately in items from the antebellum period (only about 25% of the author, date attributed files pre-date the Civil War, but 35% of items containing her husband are antebellum).

Scanning the KWIC reveals that uses of her husband are extremely negative.  Modifiers for husband includ

Preceding words  are against, abandoned, fought, ill-treated, cruelty, fear, injuries, subjected.  However, words like consent, death, property, protection invoke a second connotation, that of a legal discourse.

Looking at the lines containing her husband  I coded by searching for the following words under categories death, negative connotations, the law, property, coverture, family, and rights.  (table 3) then sorted chronologically. 

table 3 words used to code each category

Distinctive uses around coverture are most prominent in both periods, as we might expect, especially since I searched for the highest number of terms under that category.  Chronological shifts occurred though in terms of the sense of rights (both in general and in terms of specific rights)  in antebellum era items which seems to fade away during the postbellum period, giving the legal discourse more prominence.

Data Viz #5 NB 1886 is the date of publication for Volume 3 HWS. The 38 items with that date are chapter headings, appendix, and Isabella Beecher Hooker's memoir).   I'm curious about the year 1850 so I pull up those items.

Data Viz #5 her husband weight by total number of items containing per year

Mary Upton Ferrin's ADDRESS TO THE JUDICIARY COMMITTEE OF THE MASSACHUSETTS LEGISLATURE concerned the property rights of married women and the right of divorce.  About half of the instances of "her husband derive from the anecdotes Mrs. Ferrrin used throughout her appeal.

A similar argument was made in an Address to the Women of Ohio, 1850 by Jane Elizabeth Hitchcock Jones  who also provided this anecdote "In Hindostan, the evidence of woman is not received in a court of justice. The Hindu wife, when her husband dies, must yield implicit obedience to the oldest son." This focus on married wmen's property rights is what I'd expect to find in 1850.  I'm not quite so clear however about how her husband would fit into the rhetoric of 1869.  The Speech of Anna Dickinson to the Chicago Woman Suffrage Convention, 1869, in which she claims to have been exiting the hall due to the stifling heat, only to be called back to the platform by dissent voiced as follows

"that woman is abundantly represented by some man of her family; that when a woman lifts herself up in opposition against her husband, she lifts herself up, if I properly and rightly understood the declaration, against God; that the inspired assertion is that the husband is the head of the wife."

In this case, the instance of her husband is a sort of rhetorical trope attributed to the opposition to woman suffrage.  It is quite interesting through in terms of my analysis of male in the male-authored items, in which I noted in the 1850s a strong religious slant to items in History of Woman Suffrage. 

Virginia Louisa Minor, who would go on to be plaintiff in the landmark Supreme Court Decision Minor v Happersett, gave an address in 1869 to a suffrage convention in which she offered a Nativist/ xenophobic rationale for woman suffrage 

is it presumable that a foreign citizen is intended to be placed higher than one born on our soil? Under our Constitution and laws, woman is a naturalized citizen with her husband.  There are men in this town to-day, to my certain knowledge, who have had this boon of citizenship thrust upon them, who scorned the name, and who freely claimed allegiance to a foreign power.

Finally Elizabeth Cady Stanton offered an example of a widow who carried on her husband's business successfully.    

From data viz #6 Elizabeth Cady Stanton is a very frequent user of her husband but then again she is a very frequent contributor to the History of Woman Suffrage. By the number of items she authored contain her husband are proportionate to the percentage of items she contributed.  In order to determine whether her usage is skewed in terms of words count, I  need to hand separate the files attributed to her to determine word count, which I haven't done yet.  However what I can do is compare her use of her husband over time to see if it shifted.  I suspect that her husband reflects the anecdotes used to personalize the struggle for woman's rights and suffrage.

Data Viz #6 her husband by author not corrected by corpus contribution

In an 1854 address to the New York State Legislature, advocating for married women's rights, Stanton ran the gamut from invoking the slavey analogy to implying that coverture encouraged women's immoral behavior. 

"The wife who inherits no property holds about the same legal position
that does the slave on the Southern plantation. She can own nothing,
sell nothing. She has no right even to the wages she earns; her person,
her time, her services are the property of another. She can not testify,
in many cases, against her husband. She can get no redress for wrongs
in her own name in any court of justice. She can neither sue nor be sued.
She is not held morally responsible for any crime committed in the presence
of her husband, so completely is her very existence supposed by the
law to be merged in that of another. Think of it; your wives may be thieves,
libelers, burglars, incendiaries, and for crimes like these they are not held
amenable to the laws of the land, if they but commit them in your dread
presence. For them, alas! there is no higher law than the will of man."

She also invoked a xenophobic comparison between the status of widows in the United States to the practice of sati

"Man has ever manifested a wish that the world should indeed be a blank to the companion whom he leaves behind him. The Hindoo makes that wish a law, and burns the widow on the funeral pyre of her husband; but the civilized man, impressed with a different view of the sacredness of life, takes a less summary mode of drawing his
beloved partner after him; he does it by the deprivation and starvation of the
flesh, and the humiliation and mortification of the spirit.

In  Appeal and Petition to the Women of New York, 1859

To the Senate and Assembly of the State of New York:
The undersigned, citizens of - - , New York, respectfully ask that you will take
measures to submit to the people an amendment of the Constitution, allowing women to
vote and hold office. And that you will enact laws securing to married women the full
and entire control of all property originally belonging to them, and of their earnings during
marriage; and making the rights of the wife over the children the same as a husband
enjoys, and the rights of a widow, as to her children, and as to the property left by herhusband, the same that a husband has in the property and over the children of his
deceased wife."

Moving to the postbellum period, Stanton's usage reflects the general shift I observed (discussed below), towards an increasingly legal applications in this period.  In a speech to the Senate Judiciary Committee in 1872

"In England, where the right to vote is based on property and not person, the  feme sole freeholder has exercised her right all along. In her earliest history we find records of decisions in courts of her right to do so, and discussions on that point by able lawyers and judges. The  feme sole voted in person; when married, her husband represented her property, and voted in her stead; and the moment the breath went out of his body, she assumed again the burden of disposing of her own income and the onerous duty of representing herself in the Government.

The last usage of her husband  in History of WomanSuffrage is by Charlotte Perkins Gilman in an 1897 speech to the NAWSA.  "the mother shall be sacrificed to her children, and incidentally to her husband" takes on additional meaning coming from the mouth of a woman denounced as both scandalous for divorcing her husband and as an unnatural mother for sending her child back to be raised by her ex-husband and his second wife.

Coming tomorrow, the husband vs her husband,  dataviz teaser below

Playing with the medatadata, which is not complete!!!!!

Educational attainment by sex of author

site of natality and death, click to go to map

post-script, Ellen asked me about diachronic shift in how demands for the vote were articulated.  Below is the distribution of equal, female, woman, woman's suffrage by VOLUME of HWS (which are only roughly chronological). This graph is not corrected for corpus size.

documenting process as project unfolds.  all results are preliminary and subject to change!
For the American Historical Association Meetings, Jan 2014 The History of Woman Suffrage (HWS) is a six volume scrapbook of sorts started by Elizabeth Cady Stanton and Susan B Anthony that retrospectively documents the eight decades and more of the movement for woman suffrage.   Although the text is long out of copyright, I am lucky to be working with the beautifully enhanced digital version from Alexander Street Press's Women and Social Movements, which comes with the most amazing metadata I've worked with so far.

As I'm not a historian of suffrage, I struggled at first to figure out what I wanted to investigate in this corpus, but fortunately for me Ellen DuBois offered her expertise and I set off, inspired by Heather Froehlich's lovely work on gender in early modern plays, to see what a data visualization and corpus linguistics approach to HWS might reveal about gender in the movement for woman suffrage.

Of the 808 individual items in HWS, author attribution exists for a little over half (55%). [of the remaining 74% are chapter or section headings, 10% are images]  Female authorship is attributed to about 2/3rd of the corpus, 1/3rd to male authors.

Please note, all my results are expressed in number of items by males and females, not the number unique male and female authors.  There are 274 unique authors (F=135, M=95).  By unique author the corpus splits 60/40 female//male.

Using the metadata from ASP, I began exploring the relationships between sex of author, genre of item, and the first controlled vocabulary subject heading.

Male authors are proportionally represented as authors of essays, under-represented as authors of  speeches, and over-represented as authors of letters

.

data viz 1

I then looked at the subjects in male authors's letters and speeches.  The ASP metadata ascribes multiple subject headings to single items.  I used the first attributed subject heading only.  This analysis shows that male authors' letters are over represented in the following subjects. Health (83%), Law/government 50%, and religion (50%).Speeches by male authors are under represented in subjects Politics (20%) Economics ((25%) and Law/government  (27%)

data viz 2

Leaving behind the ASP metadata, I moved on to a corpus linguistics approach to gender in HWS.

Subjecting the subject of woman suffrage to a gender history approach must surmount the obstacle that the term “gender” is absent from the primary sources. But this does not mean that the underlying concepts packed into the contemporary use of the term are absent. Tani Barlow has approached the history of women’s rights (including suffrage) in China by tracing the shifting Chinese language terms for woman/female. The English language does not provide such an easy entry into this project – though the terms for the political movement were numerous, shifting, and subtly different in connotation – woman suffrage, equal suffrage, votes for women, home protection, women’s suffrage, etc. Ellen Dubois  

Corpus linguistics, which finds patterns and relationships between words in a body of texts, offers one way to find gender in an English language corpus.  To get at “gender,”  I looked at gendered pairs such as man/woman, men/women, male/female his/her she/he gentlemen/ladies fathers/mothers.  Looking at the diachronic distribution of these terms,  I find several things that seem to warrant further investigation. 

graph 1

1.  spike in vol 3 for HER that needs explaining

2.  leveling off of his in 5-6 maybe because no items from men? 

3.   decrease in the use of these gendered pronouns, and while they don’t decline at same rates, perhaps what is most significant is the depersonalization of the overall discourse. This matches the argument Buhle and Buhle made about volumes 5-6  but pushes the trend back further to volume 4 

4. dip in vol 2 her is intriguing but will require KWIC and close readings to determine.  I’m also curious what the ASP metadata can tell us here does her shift in all types of entries or in specific ones? 

5.  his, her mr mrs esp his in vol 1 miss and mrs and her in vol 6 which I know from metadata has no entries attributed to men 

Also comparing items attributed to males to those attributed to females to determine if there is a gendered pattern of discourse.  Keyness, which compares the frequency of words in one text as compared to another, in this case items authored by males versus items authored by females, reveals males used terms that denote binary sex as compared to women's use of gendered pronouns and forms of address.   

Table 1

These keyness results make me realize I need to look at the diachronic distribution of terms associated with binary sex.

I also looked at collocates, words that appear juxtaposed together at greater than frequency than chance in a body of texts. The most interesting collocates so far are His wife His own Her husband Her own Now I need to determine which items contain these collocates, look at the collocates in context, and explore how they changed over time. The graph shows some interesting correlations and divergences.  

graph 2

I've written all of the above to run by people who are more expert in the history of woman suffrage.  So by all means if you have comments, leave them here, tweet @professmoravec or shoot off an email professmoravec at gmail.