I left a LOT out of that sentence and you’re right to chastise me.
This convinces me even more that language is already highly compressed, even though we can compress it more – because vagueness and misunderstandings from missing contexts are so easy to do. It’s very easy to compress language in a way that’s not mutually understood upon decompression.
OK, so I said “All humans are very similar and have similar pace of operation and similar communication needs and abilities.”
I was attempting to say two things and did so poorly:
1) Within each of the mature languages of the world and among people who are fluent in each of the languages, ON AVERAGE, the ability to minimize misunderstanding in communication is similar in range with someone else who is fluent ‘in that language*’.
So, two people of similar fluency within a language will have a similar range of misunderstandings (like we’re having now), which in information theory would be “lossy compression”.
I compressed what I wanted to say in a way that “lost content” and was unable to communicate it successfully to you. There was not enough redundancy within what I said to capture my intended meaning for proper communication with you.
*the question of whether or not we are speaking “the same language” and are fluent in “the same language” can partially fall under REGISTER – and register can switch from sentence to sentence, phrase to phrase. To “speak the same language” would include an ability to switch registers in a similar fashion. A good example of that is communication between coworkers switching from technical jargon to making a joke about the wife, switching to technical talk about a particular sport, to complaining about paychecks to back to technical jargon. Each of those would be a register shift.
It’s domain switching basically but they call it register shifting.
So by “communication needs” I meant the “need to be understood” has a similar range between communication partners. The range from desiring to be misunderstood (evil doers, manipulative, jokers) to desiring to be understood (honest people) is similar WITHIN each language.
Within a particularly language, people with similar competency abilities (fluency) have similar competency abilities.
Fluency within a particularly language is what is tied into pace and word order. If you scramble word order within a particular language and maintain the pace, the amount of entropy increase is approximately 3.5 bits regardless of the language.
I’ll find the study I got it from. I was going deeper into it in consequences to the individual communication partners and how specific this really needs to be to hold up whereas it was talking from large corpus study.
I think it’s similar in range. One would have to find a measure of success and failure per communication session between fluent users.
For example, pick any two similarly fluent communication partners speaking Spanglish and compare to any two similarly fluent communication partners negotiating meaning in law and compare to any two similarly fluent communication partners negotiating meaning in a particular IT department talking about a particular database.
Because the personality variance will likely be similar even WITHIN similarly fluent speakers of a particular register, you’ll have similar variations in attentiveness. similar amount of “having a bad day today” which should increase entropy by similar amounts beyond word order.
Perhaps much of the entropy similarities I’m thinking of can be chalked up to the similar amounts of register shifting that occurs in standard two-way human communication.
Yet we’ve both seen massive misunderstandings happening between people who speak in highly technical languages such as mathematics or physics.
Perhaps that, too is register shifting: a fan of George Lakoff in linguistics will certainly speak a different ‘language’ than a fan of Chomsky vs a fan of Pinker as they’re each working from different theoretical basis and so their glossaries are different even though they’re all within the general field of ‘linguistics’.
Ultimately, I should have left out that sentence. After all, with such a broad statement as I made, it is FAR from obvious I was speaking both to the very broad statistical averages among broad language families (and to the range of all the personalities within), and to individual communication sessions between similarly fluent speakers negotiating meaning (which is more iffy but I go more into that under
‘s comment), but NOT from any two random speakers of any fluency speaking any register.
Leaving that sentence out would have been wiser.
Even after all I said, I can still be just plain old wrong and that’s alright but it did come from somewhere.