Epistemic status: found the link, thought it was kinda interesting, mostly kicking off discussion. 

This article felt somewhat related to "Have Epistemic Conditions Always Been This Bad?", and the question of whether humanity was (briefly) in a golden age of epistemics.

Analyzing language from millions of books, the researchers found that words associated with reasoning, such as "determine" and "conclusion," rose systematically beginning in 1850, while words related to human experience such as "feel" and "believe" declined. This pattern has reversed over the past 40 years, paralleled by a shift from a collectivistic to an individualistic focus as reflected by the ratio of singular to plural pronouns such as "I"/"we."

Original paper here: https://www.pnas.org/content/118/51/e2107848118

New Comment
2 comments, sorted by Click to highlight new comments since: Today at 6:43 AM

Odd to me that "truth," "believe," and "sense" are considered "intuition" words, along with other words describing physical sensation and terms relevant to Baysian updating. From the article:

flag-words related to intuition, believing, spirituality, sapience: spirit, imagine, wisdom, wise, hunch, mind, suspicion, believe, think, trust, faith, truth, true, belief, doubt, hope, fear, life, soul, heaven, eternal, mortal, holy, god, pray, mystery, sense, feel, soft, hard, cold, hot, smell, foul, taste, sweet, bitter, hear, sound, silence, loud, see, light, dark, bright

flag words related to rationality, science, and quantification: science, technology, scientific, chemistry, chemicals, physics, medicine, model, method, fact, data, math, analysis, conclusion, limit, result, determine, transmission, assuming, system, size, unit, pressure, area, percent

To obtain suggestions for populating the belief, spirituality, and intuition cluster we also used a thesaurus algorithm available at relatedwords.org (combining search techniques such as word embedding and Concept-Net)...

[O]ur list of 5,000 most common words and the emotional ranking of words was determined in recent years and therefore reflects relatively recent language use. Still, probably the most important caveat of using book texts is that they are a biased representation of language, a bias that may change over time.

I'm skeptical of their operationalization of the "intuition/believing/spirituality/sapience" cluster, and the authors don't provide sufficient explanation or theoretical grounding for their choice. Since the entire exercise depends heavily on this aspect of the study design, and as words have multiple meanings heavily informed by context, and with all the usual worries about reproducibility, I'm reluctant to read too much into this result.

Also skeptical. I think this is tracking something more like the rise and fall of high modernism than the rise of fall of good epistemics.

New to LessWrong?