2024 - 08 August

Before the Soul Dawn - Helen Keller on Her Life Before Self-Consciousness

Some people have a strong inner voice, and some people think mostly non-verbally. People love making a big show of how shocking it is to learn that the other type of person exists.

I believe I’m in an unusual minority of people who have lived in both camps: I had nearly no regular mental chatter through high school, but developed it some time toward the end of high school or early college.

I have always assumed it was associated with my writing ability. Writing a 3-page essay in high school felt like torture¹. Even in college, I was barely required to write more than 10 pages at a time due to my engineering course schedule. It was only after I noticed how poor my writing skills were in the working world that I made a concerted effort to improve them; as I did that, I noticed that I started to vocalize whole sentences in my head. This was new. Eventually, entire conversations started happening mentally[^2]. The article by Helen Keller feels like a similar experience to transitioning from no inner voice to having mental dialogue, but turned up by several orders of magnitude. While she claims

I cannot represent more clearly than any one else the gradual and subtle changes from first impressions to abstract ideas.

in fact, she may have given us the clearest description of what it’s like to move from a world of feelings and sensations to one with words.

The Secret Paper Passing Network in Science and Beyond

The title is a bit of a misnomer… this is not as much an exposé of some shady science cabal as it is general writing and research advice:

• Write up your results often, much more than you would simply for submitting to research papers
• Pass them to as many interested friends and colleagues who might read them
• This can start a chain reaction where they will pass their ideas to you in turn

Leaky Pipelines

• When a process consists of multiple steps, each with a fractional success rate, the end distribution will be log-normal instead of normal
• Normally distributed is to additive factors as Log-normal is to multiplicative factors
• Research output is likely log-normal due to this reason: Differences between humans are often close to Normal, but research output can vary by orders of magnitude.