Friday, January 13, 2023

A new primitive language, inspired by Mesopotamic ones & by Dylan Black's I Taught ChatGPT to Invent a Language

A new primitive language, inspired by Mesopotamic ones & by Dylan Black's I Taught ChatGPT to Invent a Language https://maximumeffort.substack.com/p/i-taught-chatgpt-to-invent-a-language


My conversation with OpenAI...

> me: hello... may I create a new language with your help? I have a basic vocabulary to start.

> AI: Of course! I'd be happy to help you create a new language. What kind of vocabulary do you have to start with, and what kind of language are you trying to create?

> me: see these examples

water - u

cloud - un

rain - unu

river - um

creek - umu

man - a

boy - an

woman - anu

girl - am

embryo person - amu

food - e

cooking - en

pan - enu

milk - em

butter - emu

to be - i

to walk - in

to eat - inu

to sleep - im

to attack - imu

one - o

two - on

three - onu

four - om

five - omu

six - omu-o

seven - omu-on

eight - omu-onu

nine - omu-om

ten - oom

in/inside - uu

out/outside - uun

above - uunu

below - uum

bad - oo

good - oon

fast - oonu

slow - oom

AI: Those examples are a good start for creating a basic vocabulary for your new language. It seems that you're using a consistent pattern for creating new words by adding syllables to the end of a root word. For example, "water" is "u", "cloud" is "un", and "rain" is "unu". This is a common way to create new words in a constructed language.

It also appears that you are using a consistent pattern for creating different forms of a word. For example, "man" is "a", "woman" is "anu", "boy" is "an" and "girl" is "am". This is a good way to make the language easy to learn and use.

I would recommend that you create more words and test how they fit together in sentences, and maybe consider creating some grammar rules. You may also want to consider creating more numbers, words for different parts of speech, and so on, to make your language more expressive. Let me know if you want me to help you further, I'll be happy to assist you.

> me: and - ii

or - iin

could you please make sentences in the new language? Example to translate: one boy cooked food for two women and a man besides the river when it rained copiously

AI: Sure! Here is a possible translation of the sentence you provided into the new language you've created:

"oom an anu e en oon anu a ii um unu ii unu"


Here's how I arrived at that translation:

"one boy" = "oom an"

"cooked" = "en"

"food" = "e"

"for two women" = "oon anu"

"and a man" = "a ii"

"besides the river" = "um"

"when it rained copiously" = "unu ii unu"     <<< like humans did and do in many languages


Please note that the grammar and word order of the new language is not specified, thus this is just a way of putting the words together, but you may want to put some grammar rules in place to make the language more expressive.


---

I also tried to force things a bit...

> me: a man attacked an embryo and cooked it for the boy

> AI: "a ii amu iim an amu ii en"


Here's how I arrived at that translation:

"a man" = "a ii"

"attacked" = "iim"

"an embryo" = "an amu"

"and" = "ii"

"cooked" = "en"

"it" = "amu"

"for the boy" = "an"

[needles repetition of the previous advice]

Between 25% to 39% of the total variance in time spent playing video games can be attributed to shared genetic traits with the remainder explained by shared environmental factors (e.g., parenting and culture)

Genetic and Environmental Influences on Playing Video Games. Toqa Hassan. Media Psychology, Jan 12 2023. https://doi.org/10.1080/15213269.2023.2165504

Abstract: In this study, I use a representative survey from the German Twin Family Panel (N = 5,472) to examine the extent to which genetic and environmental factors account for time spent playing video games on personal computers and gaming consoles. Results show that genetic variation among twins explains a non-trivial amount of variation in video game play. Through ACE modeling techniques, I find that between 25% to 39% of the total variance in time spent playing video games can be attributed to shared genetic traits with the remainder explained by shared environmental factors (e.g., parenting and culture) and environmental attributes unique to individuals. This study and its findings provide a starting point for future genetic and neurological research on video game use and effects.

We are relatively pessimistic about our skills in striking an appealing conversation with a stranger and attribute the worst talking moments to ourselves

Pessimistic assessments of ability in informal conversation. Christopher Welker, Jesse Walker, Erica Boothby, Thomas Gilovich. Journal of Applied Social Psychology, January 10 2023. https://doi.org/10.1111/jasp.12957

Abstract: Conversation is one of the most common ways of establishing social connection and satisfying the need to belong. But despite spending considerable time talking to others, many people report that engaging in informal conversation with anyone other than close friends and family makes them anxious. In this research, we explored people's assessments of their conversational ability. In Studies 1a–1c, we found that people are relatively pessimistic about their skills in conversation when compared to other common activities. We also provide support for the hypothesis that this pessimism is driven by a tendency to not engage in the usual pattern of self-serving attributions when it comes to the positive and negative moments of conversations. Instead, people attribute the low points of a conversation more to themselves than to the other person (Studies 2 and 3). We discuss the origins of this attributional pattern, as well as other potential mechanisms underlying conversational pessimism, in the General Discussion.