Suppletion of "good"

For the discussion of language mechanics, grammar, vocabulary, trends, and other such linguistic topics, in english and other languages.

Moderators: gmalivuk, Moderators General, Prelates

User avatar
Eebster the Great
Posts: 2475
Joined: Mon Nov 10, 2008 12:58 am UTC

Suppletion of "good"

Postby Eebster the Great » Tue Nov 25, 2014 5:21 am UTC

Why are certain words like "be," "good," and "I" irregular in so many languages? I find it hard to believe it's just frequency of usage. Wikipedia claims "good" is often subject to suppletion, for instance, but does not conjecture why.

User avatar
azule
Saved
Posts: 2114
Joined: Mon Jul 26, 2010 9:45 pm UTC
Location: The land of the Golden Puppies and Rainbows

Re: Suppletion of "good"

Postby azule » Tue Nov 25, 2014 10:26 am UTC

The basic words of a language were born of necessity. Not converted based on rules. It would make sense for them to be diverse so they could be easily distinguished.

I'm sorry that I'm not perfectly addressing your subject.

My guess on "better" is that it might follow a rule, that "bett" was a synonym for "good". But I haven't done any research into it.
Image

If you read this sig, post about one arbitrary thing you did today.

I celebrate up to six arbitrary things before breakfast.
Time does drag on and on and contain spoilers. Be aware of memes.

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 25397
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: Suppletion of "good"

Postby gmalivuk » Tue Nov 25, 2014 11:25 pm UTC

azule wrote:My guess on "better" is that it might follow a rule, that "bett" was a synonym for "good". But I haven't done any research into it.
Yes, that is basically what suppletion means.

I don't know why particular words would be especially prone to it across different languages, but within a particular language I expect it's a matter of people perceiving less and less of a distinction over time between two words, until at some point they're used more or less interchangeably and then the most common word in each inflection takes over that meaning entirely. "Better" may have been more common than "gooder" for whatever reason (possibly sheer coincidence--it's really unlikely two words would be exactly equally common), and so it ended up taking over the comparative role entirely.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

Derek
Posts: 2093
Joined: Wed Aug 18, 2010 4:15 am UTC

Re: Suppletion of "good"

Postby Derek » Wed Nov 26, 2014 12:47 am UTC

I think "person-people" is a relatively transparent example of suppletion. "Person" comes from the Latin "persona", meaning person and with a regular plural. "People" comes from Latin "populus", meaning people but grammatically and syntactically singular and with a regular plural. English borrowed these words as "person-persons" and "people-peoples" with the same meaning and syntax. However, because "people" refers to a collection or group of persons, it grew to supplant "persons" as the plural of "person", and became grammatically and syntactically plural in the process. The regular plurals of both words, "persons" and "peoples" still exist in restricted use, but the general form is now "person-people".

User avatar
Eebster the Great
Posts: 2475
Joined: Mon Nov 10, 2008 12:58 am UTC

Re: Suppletion of "good"

Postby Eebster the Great » Wed Nov 26, 2014 3:38 am UTC

I understand how suppletion works. It is not particularly mysterious. The examples given here probably aren't the best anyway (person/people has a particularly complicated history, and good/better/best/well comes from three different irregular Old English words). But my question was why certain words seem to be irregular in so many different languages. The verb for "to be" seems to be irregular in a huge number of languages, including every one I have looked into so far, and it is just one example among many. Some other words are commonly but not always irregular, such as "do" (which is only irregular in English to the extent the vowel needs to change, since "doed" would be awful, but which is highly irregular in some languages). And the overwhelming majority of words are irregular in just a very few languages or in none at all. This is not the distribution you would expect due to chance. Something is promoting the irregularity of words with certain meanings across geography, time, and culture.

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 25397
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: Suppletion of "good"

Postby gmalivuk » Wed Nov 26, 2014 4:41 am UTC

I seem to recall reading somewhere that "be" is irregular in every natural language, though I couldn't possibly source that claim.

Frequency is definitely part of it (which would tend to be similar across languages), but other things have an influence as well, such as analogy with similar-sounding words (which obviously wouldn't be similar across a large range of languages). People often use "snuck" as the past tense of "sneak" not because that's a fossilized old form of a common word, but because it's sufficiently similar to a group of vowel-changing irregular verbs. And then there's my baseless suppletion hypothesis above about meaning drift among near-synonyms, which would also tend to be more restricted to one language or family.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

User avatar
azule
Saved
Posts: 2114
Joined: Mon Jul 26, 2010 9:45 pm UTC
Location: The land of the Golden Puppies and Rainbows

Re: Suppletion of "good"

Postby azule » Wed Nov 26, 2014 6:52 am UTC

Eebster the Great wrote:And the overwhelming majority of words are irregular in just a very few languages or in none at all. This is not the distribution you would expect due to chance. Something is promoting the irregularity of words with certain meanings across geography, time, and culture.
Unless I miss the point, humans seek to create patterns where none might exist. You look for things to be more regular, while it's perfectly natural for it not to be.

As a language matures, more words are added along with rules governing those words. That's not true at the beginning.

I think an important thing to remember is that languages, even now, are not perfectly segregated entities. They are influenced by their neighbor languages. This means that these base words might not actually have been born by the same speakers and might actually be from "different languages" even though their etymology proposes they're from the same origin.

Why is this true for all languages? The way I imagine it is that two people will do things two different ways. Like has been said, the best of those two will be favored and that will become the one true way. (Until we change our minds again.)
Image

If you read this sig, post about one arbitrary thing you did today.

I celebrate up to six arbitrary things before breakfast.
Time does drag on and on and contain spoilers. Be aware of memes.

Aiwendil
Posts: 270
Joined: Thu Apr 07, 2011 8:53 pm UTC
Contact:

Re: Suppletion of "good"

Postby Aiwendil » Wed Nov 26, 2014 11:52 pm UTC

Suppletion requires the existence of two or more synonyms or at least semantically related words. So perhaps the words that seem to be suppleted in many languages are those that tend to have many synonyms or semantically related words? For instance, a lot of languages have multiple words meaning 'be' - to name a few, Spanish has the ser/estar distinction, Old English had beon/wesan (whence our suppletive verb), and even modern English has words like 'get' and 'become' that cover close semantic territory. Are frequently suppletive words those that correspond to densely-populated semantic regions?

Valdeut
Posts: 8
Joined: Wed Feb 26, 2014 7:14 pm UTC

Re: Suppletion of "good"

Postby Valdeut » Thu Nov 27, 2014 12:07 am UTC

Eebster the Great wrote:Why are certain words like "be," "good," and "I" irregular in so many languages? I find it hard to believe it's just frequency of usage. Wikipedia claims "good" is often subject to suppletion, for instance, but does not conjecture why.
If the question is "why are 'be', 'good', and 'I' so likely to be irregular, while other words such as 'linguistics' or 'forum' are not?", I think frequency of usage goes a long way to explain this. Words that aren't frequently used can very well develop some irregularities (especially irregularities due to sound changes) but they don't have the same ability to sustain irregularities over time. The basic assumption when we learn a new word is that it can be regularly inflected so it takes a certain level of exposure to all the different forms to break this assumption.

There may be a slightly different phenomenon going on as well, although I'm not sure how well I can explain this. Even if a language has a verbal copula (like the english "to be"), this verb may still in many ways behave differently from other verbs. So for the reason that it's not really a normal verb, it may not be subject to the same analogical pressure as other verbs. Since it's syntactically kind of its own thing, it's not obvious that it shouldn't be morphologically as well, if that makes sense. For the same reason, pronouns are not subject to that much pressure to decline like nouns (although nouns and pronouns certainly do borrow inflection from each other) although there may be some analogical pressure among pronouns.

Both of these phenomena only explain why infrequent words (and words from large word classes) lose their irregularities. A different question is of course why irregularities develop in the first place.

Suppletion is not the only source of irregularities. Sound changes can really wreak havoc to a once regular system. English am and is are not suppletive forms (in relationship to each other, that is), they can in fact be derived via sound changes from the regular conjugation of the same Proto-Indo-European root. But as far as Modern English is concerned, they don't look related at all. The same sound changes obviously applied to other less frequent words but "to be" was common and functionally different enough to sustain the irregularities that developed.

Frequent and functionally special words are also more likely to be able to maintain unusual or archaic morphology, or even maintain grammatical distinctions not found in other words. The english subject–object distinction in pronouns is a good example, as is the distinct first person singular form "am". And the source of english "to be" and "to do" were special already in Proto-Germanic in that they maintaned the old Proto-Indo-European athematic conjugation (which can be seen in the /m/ och "am"), whereas all other Proto-Germanic verbs had shifted to the thematic conjugation. Again, this is irregularity by conservatism.

Then there is obviously true suppletion such as be–is, good–better–well etc., where the different forms are historically from different roots. I haven't checked if there's any litterature on the subject. But my main hypothesis for how this develops is probably quite similar to gmalivuk's above. There is a general tendency for languages to dislike perfect synonyms, so in the case that there are two common words that come close to being perfectly interchangable (due to semantic drift or borrowing), speakers will tend to distribute the two words over different uses. The most common solution is probably to distribute the two words over different meaning, so that a new semantic distinction is developed. But another solution is to distribute the two words over different inflected forms so that they merge into one paradigm.

Another explanation is that sound changes and morphological changes can create a pressure for common words to borrow forms. Such changes can cause two forms of a word to become too similar. In order to maintain a grammatical distinction, speakers may start to use another word in some places. Add to this that common words are often short and phonologically weak, so they are much more susceptible to being "worn down" by sound changes.

Valdeut
Posts: 8
Joined: Wed Feb 26, 2014 7:14 pm UTC

Re: Suppletion of "good"

Postby Valdeut » Thu Nov 27, 2014 12:42 am UTC

About good–better–best, this example of suppletion goes back to Proto-Germanic, and there is hard to see any good reason for it. There were probably two different words at one point, but the "missing" forms are not attested in any Germanic language. We can reconstruct an adjective *bataz that gave the comparative *batizô and the superlative *batistaz and it is likely that this is the older word meaning 'good'. But not even the oldest attested Germanic languages has preserved the positive form.

The adjective *gōdaz may very well have had a regular comparative and superlative form at one point but again, these are not attested. From comparison with other IE-languages, it is likely that this adjective originally meant ’suiting’ or something similar but that it shifted to meaning ’good’ in general. Then, for whatever reason, these two now synonymous adjectives merged into one paradigm.

Interestingly, Swedish has borrowed the word bra and incorporated it into the paradigm bra–bättre–bäst. It replaced the native god which is now mostly used describing good quality in certain domains (such as taste and morality). But through analogy, this new more restricted use of god actually developed a full paradigm: god–godare–godast.

The word bra (earlier braf, or brav in modern spelling) is related to English brave and borrowed from Low Germanic, which in turn borrowed from French, which borrowed in from Italian bravo. The origin of the Italian word is apparently unclear.

gmalivuk wrote:I seem to recall reading somewhere that "be" is irregular in every natural language, though I couldn't possibly source that claim.
I don't think that's quite true. I believe the Quechuan languages have copulas that are completely regular verbs, for example, with the caveat that they may use zero copula in the present tense. I'm sure there are more examples.

The copula is probably the most likely verb to be irregular, however. As far as I understand, the verb-stem i- ’to be’ is the only irregular verb in Turkish. It's only used in some tenses and moods, with others being supplied by the stem ol- (also meaning 'to become'). Apart from this one stem-suppletion, it is conjugated regularly, though.

Also, not every language has a verb for "to be", some languages use other strategies such as a zero copula, noun-suffixes, special pronouns and stative verbs. Many languages use more than one strategy.

User avatar
Lazar
Landed Gentry
Posts: 2151
Joined: Tue Dec 29, 2009 11:49 pm UTC
Location: Massachusetts

Re: Suppletion of "good"

Postby Lazar » Fri Jan 08, 2016 4:22 pm UTC

Speaking of suppletion – I'm practicing Romance irregular verbs, and I made a little chart showing the different stems used for go:

Screen Shot 2016-01-08 at 11.18.16 AM.png

Green coming from Latin īre, red from vadēre, yellow from the fu- stem of esse, and blue and purple, as best I can tell, from a murky etymological stew involving ambulāre and some other things. I used one representative form for each tense, but in some cases there are different stems within a single tense.
And remember, my friends, future events such as these will affect you in the future.

User avatar
Quizatzhaderac
Posts: 1485
Joined: Sun Oct 19, 2008 5:28 pm UTC
Location: Space Florida

Re: Suppletion of "good"

Postby Quizatzhaderac » Wed Jun 29, 2016 9:42 pm UTC

I have another conjectural explination,

These words tend to be very vague; that is have a low information content. For example many adjectives have a good/bad connotation, which means there's basically a thousand more specific alternatives.

The information from the specific from good/better/best/well is relatively more important. For example, consider the linguistic shift:
Before the civil war, the United States were X
After the civil war, the United States was X
The presence of the copula doesn't actually provide much information; that's why we can understand cavemen when they say "Fire bad!"
The thing about recursion problems is that they tend to contain other recursion problems.

User avatar
Eebster the Great
Posts: 2475
Joined: Mon Nov 10, 2008 12:58 am UTC

Re: Suppletion of "good"

Postby Eebster the Great » Thu Jun 30, 2016 12:13 am UTC

That's actually a good point. It obviously doesn't apply to all cases ("I," "me," "we," and "us" have very precise meanings), but it does for the verbs and adjectives in the OP at least. "Be" and "do" have close to the minimum possible informational content," while "go," "good," and "bad" are still pretty vague and have the tendency to take on even more diverse meanings than their original, restricted sense would permit. Conversely, languages without words for these terms (e.g. Latin lacks a proper word meaning "do") wouldn't be subject to that. This might explain, for instance, the confusing mess that is will/would compared to the nicely regular say/said.

User avatar
Quizatzhaderac
Posts: 1485
Joined: Sun Oct 19, 2008 5:28 pm UTC
Location: Space Florida

Re: Suppletion of "good"

Postby Quizatzhaderac » Thu Jun 30, 2016 4:38 pm UTC

Actually, I'd say I/me and we/us are also good examples of this. English generally makes the subject/object distinction through purely word order, except for a few pronouns used very, very often.

For contrast, in ASL one can "conjugate" arbitrary nouns to indicate part of speech, which matters because forms pretty much equivalent to OSV, SVO, OVS, and VOS are all possible.
The thing about recursion problems is that they tend to contain other recursion problems.

User avatar
Lazar
Landed Gentry
Posts: 2151
Joined: Tue Dec 29, 2009 11:49 pm UTC
Location: Massachusetts

Re: Suppletion of "good"

Postby Lazar » Thu Jun 30, 2016 5:05 pm UTC

In fact the suppletions in "I/me" and "we/us" go all the way back to Proto-Indo-European.
And remember, my friends, future events such as these will affect you in the future.

User avatar
Eebster the Great
Posts: 2475
Joined: Mon Nov 10, 2008 12:58 am UTC

Re: Suppletion of "good"

Postby Eebster the Great » Thu Jun 30, 2016 6:46 pm UTC

Quizatzhaderac wrote:Actually, I'd say I/me and we/us are also good examples of this. English generally makes the subject/object distinction through purely word order, except for a few pronouns used very, very often.

Right, they are used frequently, but they do not have low informational content, which means they fit one hypothesis but not the other.

User avatar
Quizatzhaderac
Posts: 1485
Joined: Sun Oct 19, 2008 5:28 pm UTC
Location: Space Florida

Re: Suppletion of "good"

Postby Quizatzhaderac » Thu Jun 30, 2016 7:58 pm UTC

Well, there's a difference between having a specific meaning and a high information content.

"I" means something very specific in context, but it's commonness undermines it's information value. Consider what would happen if one dropped the subjects from sentences, how many statements about oneself would still be understandable? Woke up, had breakfast, went to work, and argued on the internet. Just told you about my day without using a subject.

Or, if we want to be mathematical: consider the set of English sentences used organized first by subject. Let's say one four start has the subject "I" and one in eight thousand has the subject "aardvark" to indicate that a sentence is one of the I-subject one is two bits of data, to indicate a sentence is one of the aardvark-subject ones is thirteen bits of data.
The thing about recursion problems is that they tend to contain other recursion problems.

User avatar
Xanthir
My HERO!!!
Posts: 5157
Joined: Tue Feb 20, 2007 12:49 am UTC
Location: The Googleplex
Contact:

Re: Suppletion of "good"

Postby Xanthir » Fri Jul 15, 2016 6:05 pm UTC

Quizatzhaderac wrote:Well, there's a difference between having a specific meaning and a high information content.

"I" means something very specific in context, but it's commonness undermines it's information value. Consider what would happen if one dropped the subjects from sentences, how many statements about oneself would still be understandable? Woke up, had breakfast, went to work, and argued on the internet. Just told you about my day without using a subject.

Or, if we want to be mathematical: consider the set of English sentences used organized first by subject. Let's say one four start has the subject "I" and one in eight thousand has the subject "aardvark" to indicate that a sentence is one of the I-subject one is two bits of data, to indicate a sentence is one of the aardvark-subject ones is thirteen bits of data.

For example, note that we can all understand Deva pretty well, despite them simply omitting subject pronouns from their sentences. It's *slightly* more difficult to understand them, but I think most of that is just from the novelty of the form, not any actual ambiguity.
(defun fibs (n &optional (a 1) (b 1)) (take n (unfold '+ a b)))

User avatar
Eebster the Great
Posts: 2475
Joined: Mon Nov 10, 2008 12:58 am UTC

Re: Suppletion of "good"

Postby Eebster the Great » Fri Jul 15, 2016 11:02 pm UTC

Deva? Like the Hindu gods? I'm not sure I can understand them.

EDIT: Wow, where did my entire last post in this thread go? I had a whole post about how the information content of "I" is pretty high and sentences are only clear without it when already given context and deliberately constructed to be so.

If, out of context, I give you a sentence like "took my time getting here," it's clear the subject must be "I," because nobody else can really take "my" time in that sense. But if I give a sentence like "got here," there's really no way if you could know if "I got here" or "he got here" or "they got here" or whatever, because the subject isn't specified. It isn't implied or obvious in the vast majority of sentences. However, once the subject "I" is given, it is completely unambiguous, because there is only ever one subject that is I. Some nouns can be fairly vague, as can the pronoun "you," but "I" is as specific as it is possible to be.

That's what I mean when I say it has high information content. I think you can also say that the word is pretty unpredictable in sentences. Usually, as someone is speaking, you can't tell that the next word is going to be "I" from the previous words. It is possible that you can tell that a previous word should have been "I" based on the following words, but that's not usually how language works.

User avatar
Lazar
Landed Gentry
Posts: 2151
Joined: Tue Dec 29, 2009 11:49 pm UTC
Location: Massachusetts

Re: Suppletion of "good"

Postby Lazar » Fri Jul 15, 2016 11:29 pm UTC

Deva's a poster here who has a distinctive writing style.
And remember, my friends, future events such as these will affect you in the future.

User avatar
Deva
Has suggestions for the murderers out there.
Posts: 1902
Joined: Sat Feb 26, 2011 5:18 am UTC

Re: Suppletion of "good"

Postby Deva » Fri Jul 15, 2016 11:37 pm UTC

Waves.
Changes its form depending on the observer.

User avatar
Quizatzhaderac
Posts: 1485
Joined: Sun Oct 19, 2008 5:28 pm UTC
Location: Space Florida

Re: Suppletion of "good"

Postby Quizatzhaderac » Mon Jul 18, 2016 6:04 pm UTC

My point was that the "decorative" information becomes more important with word that have low "core" information.

"Decorative information" being what is typical accomplished through conjugation or word order. For example a regular adjective may be decorated with 'er', 'est', or 'ly'. In English nouns are decorated with number (s) and/or possession ('s), and sometimes with gender. In ASL, nouns are decorated with subject/object distinction.
Eebster the Great wrote: I had a whole post about how the information content of "I" is pretty high and sentences are only clear without it when already given context and deliberately constructed to be so.
Well from the perspective of Huffman encoding: "I 's frequency makes it (by mathematical necessity) low information content.
If, out of context,
.....
if I give a sentence like "got here," there's really no way if you could know if "I got here" or "he got here" or "they got here" or whatever, because the subject isn't specified.
......
but "I" is as specific as it is possible to be.
"I" tends to be specific in context. Without context it means anything capable of forming words. Without context (defining your pronouns) "...got here" is only slightly vaguer than "I got here" and "He got here".

As for what information "I" does convey, we return to my original point.

Once we strip out the decorative information, English has only one pronoun (as does ASL), but since the decorative information is of such relative importance, we split the pronoun up into 17 different words.
The thing about recursion problems is that they tend to contain other recursion problems.

User avatar
Eebster the Great
Posts: 2475
Joined: Mon Nov 10, 2008 12:58 am UTC

Re: Suppletion of "good"

Postby Eebster the Great » Mon Jul 18, 2016 6:26 pm UTC

Quizatzhaderac wrote:My point was that the "decorative" information becomes more important with word that have low "core" information.

"Decorative information" being what is typical accomplished through conjugation or word order. For example a regular adjective may be decorated with 'er', 'est', or 'ly'. In English nouns are decorated with number (s) and/or possession ('s), and sometimes with gender. In ASL, nouns are decorated with subject/object distinction.
Eebster the Great wrote: I had a whole post about how the information content of "I" is pretty high and sentences are only clear without it when already given context and deliberately constructed to be so.
Well from the perspective of Huffman encoding: "I 's frequency makes it (by mathematical necessity) low information content.

That's a different kind of information that changes depending on what people tend to talk about. By that logic, the information content in the word "Brazil" decreased dramatically during the 2014 World Cup.

If, out of context,
.....
if I give a sentence like "got here," there's really no way if you could know if "I got here" or "he got here" or "they got here" or whatever, because the subject isn't specified.
......
but "I" is as specific as it is possible to be.
"I" tends to be specific in context. Without context it means anything capable of forming words. Without context (defining your pronouns) "...got here" is only slightly vaguer than "I got here" and "He got here".

No, it doesn't. It means "the thing forming this specific sentence". It does not mean "anything," it refers to one specific person, the first person, the speaker. I don't understand your argument here.

As for what information "I" does convey, we return to my original point.

Once we strip out the decorative information, English has only one pronoun (as does ASL), but since the decorative information is of such relative importance, we split the pronoun up into 17 different words.

I don't know how you can argue that there is "one pronoun" with "decoration" when each personal pronoun comes from a completely different word and has a completely different stem, a different entry in dictionaries, and is listed differently in every source that attempts to list the commonness of words relevant to this kind of discussion. It makes say that I/me is a result of suppletion, but there was never in the history of the English language a single common personal pronoun. In most languages, there is a single highly irregular pronoun for the first person, one or two for the second person, and one to four for the third person.

User avatar
Quizatzhaderac
Posts: 1485
Joined: Sun Oct 19, 2008 5:28 pm UTC
Location: Space Florida

Re: Suppletion of "good"

Postby Quizatzhaderac » Mon Jul 18, 2016 8:07 pm UTC

Eebster the Great wrote:That's a different kind of information that changes depending on what people tend to talk about. By that logic, the information content in the word "Brazil" decreased dramatically during the 2014 World Cup.
One can use different standards of what set of sentences to comprise the linguistic universe, some of them pretty bad. But we do have expectations about what we are going to hear and language favors saying certain things more than others.
No, it doesn't. It means "the thing forming this specific sentence". It does not mean "anything," it refers to one specific person, the first person, the speaker. I don't understand your argument here.
Without context, "the thing forming this specific sentence" is anything capable of forming a sentence.

That's the nature of pronouns, their meaning changes completely with context. One doesn't get to count contextual information in with the word, because the word itself doesn't provide any context, it just references it.
I don't know how you can argue that there is "one pronoun" with "decoration" ....
I'm not trying to make any point about the number of words there actually are. There are different ways to count words, and we haven't agreed which one to use.

I was pointing out how different the pronouns are from the regular nouns, and that (by the standard of nouns) pronouns are very irregular, and supposing that the forces tend to cause good/better/well to be different also tend to keep the pronouns separate.
highly irregular pronoun
Under your reckoning, what would a "regular" pronoun be?
The thing about recursion problems is that they tend to contain other recursion problems.

User avatar
Eebster the Great
Posts: 2475
Joined: Mon Nov 10, 2008 12:58 am UTC

Re: Suppletion of "good"

Postby Eebster the Great » Tue Jul 19, 2016 12:01 am UTC

Quizatzhaderac wrote:
Eebster the Great wrote:That's a different kind of information that changes depending on what people tend to talk about. By that logic, the information content in the word "Brazil" decreased dramatically during the 2014 World Cup.
One can use different standards of what set of sentences to comprise the linguistic universe, some of them pretty bad. But we do have expectations about what we are going to hear and language favors saying certain things more than others.
No, it doesn't. It means "the thing forming this specific sentence". It does not mean "anything," it refers to one specific person, the first person, the speaker. I don't understand your argument here.
Without context, "the thing forming this specific sentence" is anything capable of forming a sentence.

That's the nature of pronouns, their meaning changes completely with context. One doesn't get to count contextual information in with the word, because the word itself doesn't provide any context, it just references it.

Pronouns are inherently relational, but that doesn't make them vague. You could make the same argument for prepositions for instance, but prepositions can have very specific meanings, it's just that their entire meaning is to specify a relationship between other words or phrases. A vague pronoun would be, for instance "something," whereas a specific pronoun would be, for instance, "which." I don't mean that "something" is confusing, that people wouldn't understand it, or that it is frequently used, I mean that it gives very little information about what it is referring to. "Which" on the other hand, as a relative pronoun, always has an antecedent in the same sentence, so it gives almost perfect information.

I was pointing out how different the pronouns are from the regular nouns, and that (by the standard of nouns) pronouns are very irregular, and supposing that the forces tend to cause good/better/well to be different also tend to keep the pronouns separate.
highly irregular pronoun
Under your reckoning, what would a "regular" pronoun be?

If regularity is about forms of a word following the same pattern as other words of the same part of speech in the same language, then it seems pretty clear that for instance "that/that's," which declines almost like a noun and similarly to many other pronouns, is more regular than "who/whom/whose," and that "you/your" is more regular than "I/me/my/mine/we/us/our/ours." None of these are completely regular (e.g. none take an -s in the plural), but some are certainly closer than others, taking the "regular" form in many more cases.

User avatar
Quizatzhaderac
Posts: 1485
Joined: Sun Oct 19, 2008 5:28 pm UTC
Location: Space Florida

Re: Suppletion of "good"

Postby Quizatzhaderac » Tue Jul 19, 2016 4:37 pm UTC

Eebster the Great wrote:If regularity is about forms of a word following the same pattern as other words of the same part of speech in the same language
......
None of these are completely regular (e.g. none take an -s in the plural), but some are certainly closer than others, taking the "regular" form in many more cases.
So is the universe you're using to define regularity just pronouns or pronouns and nouns?

If just pronouns, then I/me is typical in that the subject/object is indicated by two unique words. (Different words in that one can't apply a conjugation rule to get from one to another). The number of personal pronouns makes personal pronouns the typical case of general pronouns.

If by the standards of nouns, then all personal pronouns are irregular.
Pronouns are inherently relational, but that doesn't make them vague.
.....
always has an antecedent in the same sentence, so it gives almost perfect information
If I give you directions to the library I've also given you as small amount of information, I have not given you a library's worth of information. The reference to the library is very different than the library itself.

The antecedent gives the "almost perfect information", the pronoun gives a relationship to the antecedent. In order to provide all the information in the antecedent, it would actually need to repeat the antecedent.
The thing about recursion problems is that they tend to contain other recursion problems.

User avatar
Eebster the Great
Posts: 2475
Joined: Mon Nov 10, 2008 12:58 am UTC

Re: Suppletion of "good"

Postby Eebster the Great » Tue Jul 19, 2016 6:25 pm UTC

I honestly can't figure out what you're talking about. If after I use a word, the person listening to me has very precise information about what I mean that they would not have had at all without me using that word, isn't that good enough? That is, if I give a sentence lacking the word X and it is very confusing because one key piece of information is missing, but if I give the same sentence including X it is completely clear, then have I not added a lot of information? On the other hand, if by adding X I merely rendered the sentence syntactically correct without clarifying its meaning, haven't I added very little information?

User avatar
Quizatzhaderac
Posts: 1485
Joined: Sun Oct 19, 2008 5:28 pm UTC
Location: Space Florida

Re: Suppletion of "good"

Postby Quizatzhaderac » Tue Jul 19, 2016 10:15 pm UTC

If after I use a word, the person listening to me has very precise information about what I mean that they would not have had at all without me using that word, isn't that good enough?
Yes, but that's entirely not what pronouns do. Pronouns establish a relationship to information you already have. If you don't already have that information, the pronoun is undefined.
That is, if I give a sentence lacking the word X and it is very confusing because one key piece of information is missing, but if I give the same sentence including X it is completely clear, then have I not added a lot of information?
Which provides more information?
"which was very difficult."
"Something was very difficult"
They both provide the same information, expect maybe the "which" one could be said to imply the speaker knows exactly what the something is. We wouldn't use "which" like this because it's relational, and that fragment is clearly missing what the "which" is relating to.

"He had to leave, which was very difficult" conveys more information.
There is no sensible thing to do but contribute the extra information to the extra words.

The information "Which" provides is to join "He had to leave" to "Something was very difficult". This information is tied to the nature of the word "which".

The word "which" does not provide the information "He had to leave", it merely references it, otherwise we wouldn't need the "He has to leave" for the sentence to make sense.
The thing about recursion problems is that they tend to contain other recursion problems.

User avatar
Eebster the Great
Posts: 2475
Joined: Mon Nov 10, 2008 12:58 am UTC

Re: Suppletion of "good"

Postby Eebster the Great » Wed Jul 20, 2016 2:17 am UTC

Quizatzhaderac wrote:
If after I use a word, the person listening to me has very precise information about what I mean that they would not have had at all without me using that word, isn't that good enough?
Yes, but that's entirely not what pronouns do. Pronouns establish a relationship to information you already have. If you don't already have that information, the pronoun is undefined.

Well that's how relative pronouns are used. Clearly this is generally more specific than, for instance, demonstrative pronouns, which can be quite ambiguous. You are still referring to something with a demonstrative pronoun, but it is not always completely specified to which thing you are referring. They give less information about the identity of the referent, forcing you to guess more about which it is based on ambiguous context.

That is, if I give a sentence lacking the word X and it is very confusing because one key piece of information is missing, but if I give the same sentence including X it is completely clear, then have I not added a lot of information?
Which provides more information?
"which was very difficult."
"Something was very difficult"
They both provide the same information, expect maybe the "which" one could be said to imply the speaker knows exactly what the something is. We wouldn't use "which" like this because it's relational, and that fragment is clearly missing what the "which" is relating to.

"Which was very difficult" is not a complete sentence. Taken to be a complete sentence, it is simply syntactically incorrect. On the other hand "that was very difficult" is a completely correct sentence which is obviously quite imprecise.

"He had to leave, which was very difficult" conveys more information.
There is no sensible thing to do but contribute the extra information to the extra words.

The information "Which" provides is to join "He had to leave" to "Something was very difficult". This information is tied to the nature of the word "which".

That's literally my point. The "information" is "tied to the word." The word "which," by its "nature" gives more information.

The word "which" does not provide the information "He had to leave", it merely references it, otherwise we wouldn't need the "He has to leave" for the sentence to make sense.

No, it doesn't give the information of the antecedent, it gives referential information to the antecedent. It specifies exactly which noun, having already been used, you are reusing.

User avatar
Quizatzhaderac
Posts: 1485
Joined: Sun Oct 19, 2008 5:28 pm UTC
Location: Space Florida

Re: Suppletion of "good"

Postby Quizatzhaderac » Wed Jul 20, 2016 6:30 pm UTC

Eebster the Great wrote:
The word "which" does not provide the information "He had to leave", it merely references it, otherwise we wouldn't need the "He has to leave" for the sentence to make sense.

No, it doesn't give the information of the antecedent, it gives referential information to the antecedent. It specifies exactly which noun, having already been used, you are reusing.
So you'll admit that we do not get to count the information in the referent as provided by the reference?

Pronouns do provide real information in the relationships they imply. But that relational information (divorced from the informational content of the referent), is typically only one or two bits of information.

If I describe Knuckles as "He" I'm providing a link to whatever clause the pronoun appears in. When I describe him as an echidna, I indicate that he has a spine, is warm blooded, has hair, was hatched from an egg, that he nursed from his mother, et cetra, et cetra.
The thing about recursion problems is that they tend to contain other recursion problems.

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 25397
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: Suppletion of "good"

Postby gmalivuk » Wed Jul 20, 2016 9:54 pm UTC

Eebster the Great wrote:"Which was very difficult" is not a complete sentence. Taken to be a complete sentence, it is simply syntactically incorrect. On the other hand "that was very difficult" is a completely correct sentence which is obviously quite imprecise.

Both are unclear when taken alone. Both become clear when uttered after "He had to leave":

He had to leave, which was very difficult.
He had to leave. That was very difficult.

It seems to me that "which" and "that" give identical information in the two utterances. I don't think it's important whether the second half of each one is technically a complete sentence, because the information of that half doesn't depend on sentencehood. If you like, we can use two speakers:

A: He had to leave.
B: Which was very difficult.

A: He had to leave.
B: That was very difficult.

Is the informational content of B's utterance at all different between the two cases?
---
All "which" does is provide a connection to other information. "http://forums.xkcd.com/viewtopic.php?f=25&t=110492" provides a connection to all the information on this page, but the informational content of "http://forums.xkcd.com/viewtopic.php?f=25&t=110492" is quite a lot lower than that of this whole page.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

User avatar
Eebster the Great
Posts: 2475
Joined: Mon Nov 10, 2008 12:58 am UTC

Re: Suppletion of "good"

Postby Eebster the Great » Thu Jul 21, 2016 3:31 am UTC

gmalivuk wrote:
Eebster the Great wrote:"Which was very difficult" is not a complete sentence. Taken to be a complete sentence, it is simply syntactically incorrect. On the other hand "that was very difficult" is a completely correct sentence which is obviously quite imprecise.

Both are unclear when taken alone. Both become clear when uttered after "He had to leave":

He had to leave, which was very difficult.
He had to leave. That was very difficult.

It seems to me that "which" and "that" give identical information in the two utterances. I don't think it's important whether the second half of each one is technically a complete sentence, because the information of that half doesn't depend on sentencehood. If you like, we can use two speakers:

In my mind at least, the two examples do not give identical information. "Which" or "that" used as a relative pronoun are highly specific. It's not just about which relationship they specify, it's about using a relative pronoun at all giving a lot of information as compared to using any other part of speech. And the relative pronoun "that" has nearly identical meaning to the relative pronoun "which."

But the case I was referencing was using "that" as a demonstrative pronoun, which is less specific. There are other kinds of context you can use to clarify its meaning, usually including verbal context like you gave here, but sometimes situational, even gestural. So I guess the amount of information it provides really depends on the specific case. But in any case, the maximum information it could provide is its use as a relative pronoun, making "which" effectively an upper limit of "that".

Either way, the two are declined identically.

All "which" does is provide a connection to other information. "http://forums.xkcd.com/viewtopic.php?f=25&t=110492" provides a connection to all the information on this page, but the informational content of "http://forums.xkcd.com/viewtopic.php?f=25&t=110492" is quite a lot lower than that of this whole page.

Certainly a pronoun provides less information than its antecedent; I agree with that. However, that does not mean every pronoun provides an equal amount of information. A pronoun that exactly specifies an antecedent gives perfect referential information (even if it's still merely referential), whereas a pronoun that inexactly specifies an antecedent gives imperfect referential information.

For comparison, the above URL certainly gives more information than the relative URL /viewtopic.php?f=25&t=110492.

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 25397
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: Suppletion of "good"

Postby gmalivuk » Thu Jul 21, 2016 7:17 am UTC

In my example, "That" is demonstrative, not relative.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)


Return to “Language/Linguistics”

Who is online

Users browsing this forum: No registered users and 3 guests