BIG PICTURE: My research revolves around a pattern in which visual artists who are seminal (maintain a lifelong making practice, are prolific and are recognized in the art world) also tend to have a writing practice. My suspicion is that visual artists’ writing, like expressive-experimental writing, has some form of psychological resilience and/or realization function. Additionally writing may provide some kind of cognitive processing needed to counter balance or deepen the largely bodily practice of art. Relative to the expressive-experimental writing which is a form of disclosure there seem to be five prevalent notions that suggest why written disclosure may function beneficially for the writer—inhibition theory, cognitive processing theory, self-regulation, social integration and an exposure model (Frattaroli, 2006). My guess is that it involves all to varying degrees.
While toying with an exploration of artist generated texts, I have been dabbling with the question: In written language of female visual artists are there patterns of positioning that suggest they somehow limit themselves within the art world hierarchy? This is a rather grandiose research question likely to remain unsolved as a result of the numerous variables at play. Nonetheless it led me to review literature on gender practices in regards to language, power, stance, politeness, context, etc. Additionally, this research question was simply a grand leap into the world of computer assisted computational text analysis. Realistically, my hope was that some language differences between male and female artists would be easily recognizable after processing the samples through the text analysis software Linguistic Inquiry and Word Count, LIWC2007, which spits out a hell of a lot of numbers. I then also turned to DICTION7, which I thought might be a tad more dummy proof, neophyte scholar friendly, since it comes with some built in norm. These norms are largely built on political, journalistic and court texts (speeches, manuscripts, etc). This population has historically been male dominated; therefore DICTION’s norms largely reflect patterns culturally ascribed to as male practices. Between the two-software analysis two categories were noteworthy in the visual artists’ data—INSISTENCE and CERTAINTY.
That these variables were of note fits tightly to my lived understanding that much of art making manifests as reactionary. By this I mean that somehow a particular developmental adversity, non- or anti-hierarchical notion, or situational chaffing unsettles or resonates the artist to the point that they engage a form of appraisal, judgment or questioning, that is then translated via artistic production addressing, calling attention to, resolving, acting against or support of the artist’s viewpoint or experience. Linguistic insistence and certainty expressed in writing would also appear to be behaviors in which appraisal/judgment is expressed, in which the writer takes a stance. Because most art is reactionary, a form of stance taking and is what captivates the mind and energy of an artist, I expect stance taking, even tentative stance taking, to be statistically evident in their writings.
QUESTIONS (this week’s set)
A visual stance (art making) involves a bodily response and use of materials and space. A cognitive stance is expressed through linguistic space orally or in text. Will these be mirrored in artifacts and text emerging from a single author/maker? Will this mirroring be evident in most seminal visual artists’ writings?
Does strong sentiment correlate with stance taking? Is sentiment relevant to stance taking?
Does insistence and/or certainty have a comorbidity with strong sentiment? Is sentiment prevalent in artists’ writings? Is there an overarching sentiment linguistically expressed? Are there notable gender differences?
Is there a relationship or comorbidity?
[CERTAINTY + INSISTENCE < ? > Strong POSITIVE OR NEGATIVE SENTIMENT] [LOW CERTAINTY + INSISTENCE < ? >NEUTRAL SENTIMENT]
?? DEFINING SENTIMENT
Romance should never begin with sentiment. It should begin with science and end with a settlement. — Oscar Wilde, An Ideal Husband
People tend to share their views when they have an opinion. Opinions are derived from an implicit or explicit appraisal. An appraisal is the process of taking an evaluator stance. Sentiment is thought to arise from emotions or urges linked to the process and outcome of appraisals (Boiy, 2010). So does linguistic sentiment measure appraisal or affect? Shrug (It is dependent on your literature review). Are they the same or different? Shrug (I lean toward different, yet they are often equally measured in terms of positive, negative, neutral, ummm, sentiment). What is the intent of sentiment analysis? What does it measure? Does it characterize opinionated or evaluative aspects of natural language? Kenneth Bloom and cohorts (2007) suggest that a linguistic appraisal is a textual unit asserting an evaluative stance towards some target. Provalis in their literature on WordStat7 defines sentiment analysis as the “application of analytic techniques for the identification of subjective opinions in text data” But whose techniques are being applied? Provalis also states that out of the box systems, dictionaries, tend to be domain dependent based on the intent during their development and that the developers domains may connate positive, negative, neutral to a very specific use of language that might not equally apply to other populations being studied. They note the example that “freezing” is positively associated with refrigerators and negatively associated with software. They suggest, strongly, that each study using WordStat7’s built in sentiment analysis should have the dictionary validated/tweaked based on the data population being studied. Sigh. WordStat also provides access to Loughran and McDonald Financial Sentiment Dictionary and Lexicoder Sentiment Dictionary (LSD) by Young and Soroka (2011). LSD is a hybrid compilation of several different dictionaries and rule sets from multiple domains with the intent of being less domain specific. LIWC, also available as an add on to WordStat, relies less on content words and more on function (style) words which are less domain specific. LIWC also labels the positive and negative categories with the word emotion—positive emotion.
If any exist, what are the relational patterns between sentiment and certainty and insistence? Are there notable gender differences? Does it actually seem relevant to my larger questions about the function of writing for visual artists or is this just a nice rabbit trail that will expand my general psycholinguistic understanding and methodologies?
SOFTWARE TOOLS and DICTIONARIES NEOPHYTELY EMPLOYED
QDA Miner and WordStat7 using the built in Sentiment Dictionary and the additional dictionaries from LSD, RID and LIWC.
THINKING OUTLOUD, PEN TO PAGE:
Pennebaker (1992) relative to the inhibition theory suggests that those with low inhibition constraints exhibit greater benefit from written experimental disclosure. Does lower inhibitions correlate with creativity research that suggests that seminal creators are less compelled to fully adhere to social conventions. In creativity literature what sifts to the surface is that creators, will have or have experienced developmental adversity (loss, illness, psycopatholgy, etc), neurological sensitivity (loose-sensory gating which correlates with non-heiarchical thinking), personality dispositions that tend towards oppositional, situational chaffing. It is believed that any combination of these lowers resistance for the individual to push against social constraints. In my mind this correlates with Pennebaker’s low inhibitional constraints. If Pennebaker is correct that those low in inhibiting constraints benefit from expressive writing, then it would follow that artists are likely to benefit from expressive writing.
(stopping mid thought)…0