3 ms·

> no subjunctive tenses > no tones etc. According to this view a perfect language would have... what? So are we in the same game as in computer language, with

by gbog 11y ago

> no subjunctive tenses > no tones etc. According to this view a perfect language would have... what? So are we in the same game as in computer language, with this discussion about how many features should we allow? My view on this may be uncommon: Less features for computer language migth be better in some cases, but more features in human languages is always better, because the human brain will make the best out of it. Taking a simple case: two idioms with the same meaning is bad in python, should be removed if possible. Two words with the same meaning in English (e.g. to use/to utilise, tired/fatigued) is good, because both words will fork away and each carry subtle differences that will be powerful. English is highly unsuitable, but not for this reason.

netheril96 11y ago
You are conflating language constructs with words. More words may be fine, but more "features" are not. Frankly, those "features" are more like bugs. What on earth is the usage of masculine or feminine nouns, or irregular past tense/participle, or to/ing difference?
- schoen 11y ago
  There's a theory that languages have sometimes evolved extra redundancy to help with understanding in noisy environments (in every sense of noisy, but especially if it's physically difficult to hear someone). This could be viewed as akin to a parity bit in telecommunications, where it's deliberately redundant to increase the chance of detecting errors on a noisy channel, or with other ECC methods of correcting them. Noun gender might conceivably work that way some of the time, for example if there are nouns that otherwise sound fairly similar but are distinguished by their genders; if you heard an article or an adjective that agreed with the noun, you could get an extra clue about what it was. An example that I thought of in Portuguese is ato/ata, two closely etymologically-related words with different grammatical genders. The former means 'act, deed', while the latter means 'record, minutes (of a meeting)'. If there were a context in which these could be confused when spoken, surrounding words would likely reveal which one had been said. I'm not sure whether this is commonly understood to be a likely reason that languages actually have systems of agreement. Edit: another example of an agreement mechanism that people have to spend time studying is agreement of verbs with their subjects (in various ways depending on the language -- for example Hebrew and Arabic verbs agree with their subject in gender!). This is more complex and more effort to learn than in languages that don't do this or do it less, like Chinese and Swedish (which don't even mark person or number in the verb). You can imagine that this redundancy is sometimes helpful when you mishear either the subject pronoun or the verb itself, like the different between "we want" and "she wants" or "we know" and "she knows" -- if you didn't hear for sure whether the subject pronoun was "we" or "she", having it be fully or partially marked in the verb conjugation would help you figure out which it could have been.
  - mapt 11y ago
    There are communities that aspire to degender their terminology because of pseudonyms and sensitivity to problems of the transgender community. They revert to 'They' as singular agendered personal pronoun. As this is also the word we use for multiple people, conversations about Alice and Bob fork meanings at every usage; Are we talking about Alice, about Bob, or about Alice and Bob? Even a simple sentence often has multiple occurences of personal pronouns, so you get a combinatorial explosion of ambiguity. None of https://en.wikipedia.org/wiki/Gender-specific_and_gender-neutral_pronouns#Summary https://en.wikipedia.org/wiki/Gender-specific_and_gender-neu... have adoption to the point where they would be immediately understood by 1 person in a thousand, so it's really difficult to navigate the taboo.
    - comex 11y ago
      At least it's 'only' a few pronouns - it may sound awkward now, but not so awkward that it couldn't become acceptable in a timespan of decades or less. Compare to the many languages that assign a gender to every noun, where you'd have to make up a whole new set of endings or something. (Of course, there are also plenty of languages with no grammatical role for gender whatsoever.)
    - wk_end 11y ago
      Singular "they" is incredibly widely adopted and would be and is easily understood by 1000 of 1000 native English speakers, and would be and has been understood by them for centuries: as the Wikipedia article (directly linked to by yours) says, "the singular they had emerged by the mid-14th century and is common in everyday spoken English". If we've been able to deal with this supposed combinatorial ambiguity for ~700 years, how bad could it be? If we can deal with singular and plural "you", why not singular and plural "they"? It's certainly not common, even amongst "PC" types or whatever, to want to eliminate gendered pronouns entirely; just that everyone gets to be called the pronoun of their choice (regardless of whatever genitals they were born with), and when the antecedent is of indeterminate gender, a pronoun of indeterminate or at least not a male-default gender is chosen. So Alice and Bob can still be "she" and "he" respectively if they so choose, and there's no confusion there; "they" still refers to both of them. But when a person wants to tell the story of Alice and Bob, that person is "they" or maybe "she" or sometimes "he" - just not "he" by default. Anyway, slicing up our pronoun space into male/female is entirely arbitrary. Perhaps Alice and Bob are the people chosen for these kinds of stories because our pronouns are gendered; if our language categorized people in different ways than gender, we would chose people as examples along those lines instead. And what do we do when we have a story about Adam and Bob? They're both "he"! How ever do we deal with the combinatorial explosion of ambiguity? And yet people somehow have managed to tell stories with multiple men in them for millennia... Suppose Adam is tall and Bob is short. Maybe English ought to have different pronouns for men of different heights (or hair colour, or...). It would help relieve this terrible ambiguity here. And yet we get by without this precision in our pronouns. Perhaps Adam is black and Bob white. Should they have different pronouns to distinguish them? If you say no to that and feel like that maybe might even kind of be a bad idea because of the rigid line it draws between two people created equal, well, now you know why people might be opposed to gendered pronouns. Some languages have the notion of politeness encoded in, which can often help distinguish pronouns in stories between children and adults, or royalty and their subjects. English doesn't really have this (at least not at a grammatical level), and yet somehow I don't feel my language to be particularly enfeebled without them - I can tell stories about little girls talking to their mothers, or male peasants talking to kings, just fine.
- lisivka 11y ago
  These features are allowing to build more complex sentences using less words and/or make them shorter. After translating of thousands of sentences from English to my native language for opensource projects, I found that word-to-word translations are about 30% longer in average, while translations of sentences of text are about 10% shorter in average, sometimes even 3x shorter. Direct consequence is that it is easier to understand shorter text, because of limited size of short-time memory. I.e. shorter text adds few points to IQ.
conceit 11y ago
Very helpful right now having words, that have the same meaning, to convey a different meaning.

by gbog 11y ago

netheril96 11y ago

You are conflating language constructs with words. More words may be fine, but more "features" are not. Frankly, those "features" are more like bugs. What on earth is the usage of masculine or feminine nouns, or irregular past tense/participle, or to/ing difference?

schoen 11y ago
There's a theory that languages have sometimes evolved extra redundancy to help with understanding in noisy environments (in every sense of noisy, but especially if it's physically difficult to hear someone). This could be viewed as akin to a parity bit in telecommunications, where it's deliberately redundant to increase the chance of detecting errors on a noisy channel, or with other ECC methods of correcting them. Noun gender might conceivably work that way some of the time, for example if there are nouns that otherwise sound fairly similar but are distinguished by their genders; if you heard an article or an adjective that agreed with the noun, you could get an extra clue about what it was. An example that I thought of in Portuguese is ato/ata, two closely etymologically-related words with different grammatical genders. The former means 'act, deed', while the latter means 'record, minutes (of a meeting)'. If there were a context in which these could be confused when spoken, surrounding words would likely reveal which one had been said. I'm not sure whether this is commonly understood to be a likely reason that languages actually have systems of agreement. Edit: another example of an agreement mechanism that people have to spend time studying is agreement of verbs with their subjects (in various ways depending on the language -- for example Hebrew and Arabic verbs agree with their subject in gender!). This is more complex and more effort to learn than in languages that don't do this or do it less, like Chinese and Swedish (which don't even mark person or number in the verb). You can imagine that this redundancy is sometimes helpful when you mishear either the subject pronoun or the verb itself, like the different between "we want" and "she wants" or "we know" and "she knows" -- if you didn't hear for sure whether the subject pronoun was "we" or "she", having it be fully or partially marked in the verb conjugation would help you figure out which it could have been.
- mapt 11y ago
  There are communities that aspire to degender their terminology because of pseudonyms and sensitivity to problems of the transgender community. They revert to 'They' as singular agendered personal pronoun. As this is also the word we use for multiple people, conversations about Alice and Bob fork meanings at every usage; Are we talking about Alice, about Bob, or about Alice and Bob? Even a simple sentence often has multiple occurences of personal pronouns, so you get a combinatorial explosion of ambiguity. None of https://en.wikipedia.org/wiki/Gender-specific_and_gender-neutral_pronouns#Summary https://en.wikipedia.org/wiki/Gender-specific_and_gender-neu... have adoption to the point where they would be immediately understood by 1 person in a thousand, so it's really difficult to navigate the taboo.
  - comex 11y ago
    At least it's 'only' a few pronouns - it may sound awkward now, but not so awkward that it couldn't become acceptable in a timespan of decades or less. Compare to the many languages that assign a gender to every noun, where you'd have to make up a whole new set of endings or something. (Of course, there are also plenty of languages with no grammatical role for gender whatsoever.)
  - wk_end 11y ago
    Singular "they" is incredibly widely adopted and would be and is easily understood by 1000 of 1000 native English speakers, and would be and has been understood by them for centuries: as the Wikipedia article (directly linked to by yours) says, "the singular they had emerged by the mid-14th century and is common in everyday spoken English". If we've been able to deal with this supposed combinatorial ambiguity for ~700 years, how bad could it be? If we can deal with singular and plural "you", why not singular and plural "they"? It's certainly not common, even amongst "PC" types or whatever, to want to eliminate gendered pronouns entirely; just that everyone gets to be called the pronoun of their choice (regardless of whatever genitals they were born with), and when the antecedent is of indeterminate gender, a pronoun of indeterminate or at least not a male-default gender is chosen. So Alice and Bob can still be "she" and "he" respectively if they so choose, and there's no confusion there; "they" still refers to both of them. But when a person wants to tell the story of Alice and Bob, that person is "they" or maybe "she" or sometimes "he" - just not "he" by default. Anyway, slicing up our pronoun space into male/female is entirely arbitrary. Perhaps Alice and Bob are the people chosen for these kinds of stories because our pronouns are gendered; if our language categorized people in different ways than gender, we would chose people as examples along those lines instead. And what do we do when we have a story about Adam and Bob? They're both "he"! How ever do we deal with the combinatorial explosion of ambiguity? And yet people somehow have managed to tell stories with multiple men in them for millennia... Suppose Adam is tall and Bob is short. Maybe English ought to have different pronouns for men of different heights (or hair colour, or...). It would help relieve this terrible ambiguity here. And yet we get by without this precision in our pronouns. Perhaps Adam is black and Bob white. Should they have different pronouns to distinguish them? If you say no to that and feel like that maybe might even kind of be a bad idea because of the rigid line it draws between two people created equal, well, now you know why people might be opposed to gendered pronouns. Some languages have the notion of politeness encoded in, which can often help distinguish pronouns in stories between children and adults, or royalty and their subjects. English doesn't really have this (at least not at a grammatical level), and yet somehow I don't feel my language to be particularly enfeebled without them - I can tell stories about little girls talking to their mothers, or male peasants talking to kings, just fine.
lisivka 11y ago
These features are allowing to build more complex sentences using less words and/or make them shorter. After translating of thousands of sentences from English to my native language for opensource projects, I found that word-to-word translations are about 30% longer in average, while translations of sentences of text are about 10% shorter in average, sometimes even 3x shorter. Direct consequence is that it is easier to understand shorter text, because of limited size of short-time memory. I.e. shorter text adds few points to IQ.

conceit 11y ago

Very helpful right now having words, that have the same meaning, to convey a different meaning.