Talk:Inflection

From Wikipedia, the free encyclopedia

You have new messages (last change).

What's with the new agglutination addition? I'm pretty sure that "weak inflection" is different from agglutination; things like the English third-person singular present verb suffix -s is inflection rather than agglutination, because in agglutination one affix has one meaning (so the third person, the singular, and the present would each have their own affixes). Inflection is not limited to stem changes. -Branddobbe, 11:42 PST Wed Nov 26 2003

Yeah, you're right Branddobbe, wanna fix the agglutination definition?

Contents

[edit] aint no stub

By the way I think this ain't no stub no more. Anyone disagree? Steverapaport 14:28, 1 Dec 2004 (UTC)

[edit] How is Inflection being confused with Synthesis?

(moved question from User_talk:Ish_ishwar)

Hi Ish.

You've been putting lots of flags on Inflection without ever actually discussing the confusion you point out.

If my understanding is correct, an inflected language changes nouns and/or verbs, by either agglutination or fusion, to reflect grammatical functions such as case, tense, mood, etc.

A synthetic language is a term for languages that place more than one morpheme (meaning unit, including grammatical markers), into a single word

By these definitions, all inflected languages are synthetic, though not all synthetic languages need be inflected, right?

So how (other than organizationally) is the Inflection article in need of loving attention? Can you help me make it right?

Best regards, Steverapaport 11:41, 3 Dec 2004 (UTC)


Ish ishwar reply:
Hi Steverapaport. I was hoping some of the authors would check this out at a library.
There are a number of things that I think need to be clarified. I am going to be fairly critical. The authors can decide that needs to be done.
First re: your question:
  1. Inflection is not restricted to nouns and verbs--adjectives & other grammatical categories may be inflected.
  2. One technical point: agglutinating & fusional refers mostly to the degree of segmentability of morphemes within a word. So, a language that has inflection will use inflectional morphemes. These words with inflectional morphemes may be described (in theory) as either agglutinating or fusional depending on how easily you can separate the individual morphemes.
about the Inflection entry:
  1. "inflected language" is not such good a term.
    1. Firstly, it is more common to use "inflectional language" instead of "inflected language".
    2. Secondly, using the term "inflected language" is not good due a confusion on these web pages between "inflection" and "degree of synthesis". Additionaly, "Inflectional language" has been used synonymously with "fusional language", but I dont recommend this usage. Comrie (1989: 45) has a nice discussion of this: "In place of the term fusional, one sometimes finds the term flectional, or even inflectional, used in the same sense. This is not done in the present work to avoid a potential terminological confusion: both agglutinating and fusional languages, as opposed to isolating languages, have inflections, and it is therefore misleading to use a term based on (in)flection to refer to one only of these two types. The availability of the alternate term fusional neatly solves the terminological dilemna."
  2. So the 1st sentence of the Inflection entry reads: "An Inflection or inflexion is a change of word form according to grammatical function, which occurs in inflected languages." What does the author mean by "inflected languages"? Fusional languages? Or a language that uses inflectional morphemes? If s/he means the latter, then this is partly a circular (and hence redundant) statement.
  3. 2nd paragraph. It seems that one could define inflection without recourse to the notion of degree of synthesis (i.e., isolating vs. synthetic langs). (Some may just want a clarification of what inflection is, but not really want to go into the details of morphological typology.) With this said, the paragraph is correct: all languages with inflectional morphemes will be synthetic languages.
  4. I think it would be good to point out that inflection does not include only affixation (i.e., prefixes, suffixes, infixes, circumfixes) and exclude morphological processes like reduplication & alternation.
  5. "on-the-fly"?? Does this feel to you like the author means isolating languages are more methodical or not as careless in their grammar? What does this mean?
  6. The entry mentions nothing of inflectional morphemes. Inflectional morphemes, in theory, contrast with derivational morphemes. In short, inflectional = morphemes indicating grammatical info (i.e., case, number, person, grammatical gender/word class, mood, mode, tense, aspect, other relational info); derivational = morphemes that change meaning, i.e., creating a new word from an existing word, sometimes by simply changing grammatical category (noun to verb). I think that a good discussion of inflection needs to discuss derivation.
  7. "declension" & "conjugation" are traditional grammarian terms. An introductory linguistic textbook probably wont even mention them. But, they are useful to students of European langs.
  8. "weak" & "strong". I think that these terms are not so useful. I believe they are restricted in usage to mostly Indo-European langs, perhaps mostly in historical linguistics & in tradition grammar, especially with Germanic languages (Old English, proto-Germanic) (I have seen these terms referring to PIE, too). (However, I admit that I dont know so much about the usage of these...). Maybe these terms can be mentioned in passing. But, readers should know that these terms are not so widely used out of these contexts. (Somebody should check this out)
  9. "Linguistically, the former is strictly called agglutination, and the latter is the true sense of the word inflection." Wrong. Affixation does not entail agglutinating morphology, just look most of the languages in Europe.
  10. The first part of the entry seems to discuss inflection. The second section "Inflection in various languages" seems to discuss degree of synthesis. This is the main confusion I mention in #1.2 above & on my note on Synthetic languages entry. These are two different concepts. They are related in certain ways, but nonetheless different. Some books may confuse or fail to clarify them--maybe this is why they are confused in Wikipedia? Anyway, you can see why Comrie suggests not using "flectional" to mean "fusional". [I think that somehow the term "inflectional" was mistakenly equated with "synthetic" due to examples of synthesis used in some sources. Examples of synthesis commonly use Indo-European langs which often have many inflectional suffixes (since it is very easy to get examples of these from any grammar book). So, because of these examples & also maybe due to confusing terminology mentioned by Comrie (i.e., fusional = flectional), the Wikipedia authors were misled.]
  11. Although this information in the second section is useful & good, maybe it better belongs on the Synthetic languages page. Or maybe someone should create a Degree-of-synthesis page? (maybe not?).
  12. Why dont you give your full reference of (Agirre et al, 1992)? I think it would be nice for Basque learners & others interested in these things. It is a cool fact, maybe somebody wants to check out the source. So, I suggest making a References subsection.
The above is a lot, I know. I think that the writers need to consult an introductory linguistics text. A good one is
  • Stewart, Thomas W., Jr.; & Vaillette, Nathan. (2001). Language files: Materials for an introduction to language & linguistics, (8th ed.). Columbus: Department of Linguistics, The Ohio State University.
[But dont get the 9th edition! It has a lot of confusing errors (they changed from Americanist transcription to British IPA).]
References:
  • Comrie, Bernard. (1989). Language universals and linguistic typology, (2nd ed.). Chicago: The University of Chicago Press. [Originally published (1981)].
Ish ishwar 18:59, 3 Dec 2004 (UTC)

[edit] some links

Here are some sites you might want to check out:

- Cheers! Ish ishwar 19:31, 3 Dec 2004 (UTC)

[edit] Wow

Ok, then, Ish. You've convinced me that even if I (or the other current authors) rewrote the page to clearly include all the distinctions you point out, I would lack the background to be properly clear.

You've already done most of the work in your comment above, and you obviously have the background -- would you kindly consent to improve the page yourself? I promise not to object if you throw away all my work and substitute something clearer and more correct.

Your idea for hyperlinks and references is well taken. I'll put in the Agirre reference when I find it again.

Steve

[edit] We who are doomed to fail salute you

Ok Ish, even though you could have done a better job, I've taken up the challenge

Steverapaport 17:23, 5 Dec 2004 (UTC)

Hi Steve -
So, I finally got around to this. I am not finished, but it is a start. Please improve it in any way you see fit. - Ish ishwar 11:39, 2004 Dec 22 (UTC)
I like it a lot, Ish. Glad you came in and did the necessary work. Now that you've added the sections it cries out for more content, but it's past my level of expertise. I'm just cleaning up little bits now, hope you or Mark Dingemanse or someone else comes in and adds more to the anemic sections.
Best, Steverapaport 00:11, 24 Dec 2004 (UTC)

[edit] Latin and Other Romance Languages

The bit about Latin and ablaut / umlaut is a bit confusing. The text here notes:

  • There is no Ablaut or Umlaut, and only little predictable vowel alternation, found on certain verbs where the Latin root had the phonemes /E/ or /O/.

But then when I go look at the ablaut page, I find this:

  • Latin displays ablaut in verbs such as ago (present tense), "I drive"; egi, (perfect tense), "I drove".

So which comment is correct? --- Eirikr 07:51, 6 Apr 2005 (UTC)

I've corrected that. The comment on "no Ablaut" belongs on the previous paragraph on Romance languages (which to my knowledge don't have Ablaut, or at least not a strong pattern of Ablaut inflection). Seems the Latin paragraph was added later (in the wrong place). My question would be: are some of the "irregular" vowel alternations found in Romance languages actually Ablaut carried over from Latin, or diachronical phonologically-conditioned changes of Latin vowels? I mean things like Spanish decir-digo or tener-tuve or morir-murió, not the e/ie, o/ue alternation. Also, does Romanian (often forgotten in reviews of Romance languages) have Ablaut? --Pablo D. Flores 12:40, 6 Apr 2005 (UTC)
Thanks for clearing that up, Pablo. And the new question you bring up is a good one -- the vowel shifts there in Spanish certainly seem like the ablaut talked about in strong verbs, where vowel shifts indicated tense/aspect. --- Eirikr 06:26, 7 Apr 2005 (UTC)

[edit] Japanese

Though a minor point within the scope of the whole article, might we change the wording describing Japanese as "a probable language isolate" to read "a possible language isolate"? For one, Japanese is not listed in the List of language isolates. There is also considerable debate about the relationship between Japanese and other languages, and much of this appears (at least to me) to be more to the extent of how much Japanese is related than whether or not it's related at all. Some of this debate is evident here on Wikipedia, given a look at the Japanese language and Korean language pages, and the Japanese and Korean talk pages.

Also, adjectives exhibit markedly more inflection than nouns. The description of the particles and their use to mark the case of noun phrases is accurate, but the current wording makes it sound like adjectives are as regular and un-inflected as the nouns are, which is simply not the case. Adjectives are inflected for tense ("good": yoi present, yokatta past), to mark an adverbial role (yoi mono "good thing" vs yoku suru "well do -> do (something) well"), and to mark a sort of continuation (sore ga yoi "that is good" as a complete utterance, vs sore ga yokute "that is good...", implying that another phrase is to follow). I'm not sure if it counts as inflection, but adjectives also have a noun form (yosa "goodness"), and in extreme cases, some adjectives exhibit very different forms in polite speech, with yoi changing to yoroshii. Instead of "shows a high degree of inflection on verbs, less so on adjectives and nouns," how about "shows a high degree of inflection on verbs, less so on adjectives, and least on nouns"?

Cheers, Eiríkr Útlendi 00:07, 6 January 2006 (UTC)

I've removed the mention of Japanese's being a language isolate, since (1) it's controversial, (2) it's not really relevant -slash- worth getting into, and (3) not all the other languages mentioned give their language families, so it's not like removing that statement suddenly makes the article inconsistent.
I also made the change you suggested about showing the gradations of verb inflection > adjective inflection > noun inflection.
Some of your examples look to me to suggest something more than just agglutinativity? I mean, does every adjective change -i to -roshii to become polite? (Well, I guess not, since you say "some adjectives" - but that casts doubt on the article's statement that Japanese is "extremely regular.") Please make any changes that you think appropriate in order to make the article more accurate. (Remember to be bold!) :-)
Ruakh 06:28, 6 January 2006 (UTC)
Thanks, Ruakh. I'm not an expert of much, so I hesitate to be bold in many areas without first talking it over.  :)
About Japanese and regularity, the usage of particles is indeed very regular, but the inflection of verbs and adjectives does have some exceptions. I hedged my bets about adjectives and polite forms as the only one I can bring to mind is yoi -> yoroshii, but I'm not a native speaker and there may be others I'm just not aware of. Verb-wise, the big exceptions are suru "to do", kuru to come, and to some extent iku "to go". There is also some disagreement even within Japan about what constitutes the proper potential form for the class of verbs ending in ru (the so-called Group 2 verbs), with the ending given as either rareru (identical with the passive) or reru. But aside from these, inflection is indeed awfully regular, which happily makes that aspect a bit easier to learn :), so I'm fine leaving the "extremely regular" statement as-is.
It is polite forms where things get complicated. There is some simple noun alteration for politeness where either an o- or go- is prefixed (with the go- attaching to words of Chinese origin). I assume this counts too as inflection? Also, though adjectives again have very few (possibly only the one?) polite forms, verbs undergo more common transformation in polite contexts, with iru "to be" becoming imasu (consistent with usual verb conjugation), but from there changing to gozaimasu when the speaker describes themselves or their group, and irasshaimasu when describing the party spoken to. Similarly, miru "to see" becomes mimasu, and thence haiken shimasu (compound verb, describing speaker) and goran ni narimasu (also compound, describing audience); there are numerous other examples. Is this kind of context-dependent wholesale subsititution also described as inflection, or is this something else entirely?
Cheers, Eiríkr Útlendi 16:57, 6 January 2006 (UTC)


[edit] What about Semitic languages?

Are some of them inflected? Does anyone know? —The preceding unsigned comment was added by 132.170.49.168 (talk • contribs) .

Yes, the Semitic languages are inflected. The details vary from language to language, but nouns are generally inflected for definiteness (definite/indefinite/possessed) and number (singular/plural or singular/dual/plural), and sometimes for case (nominative/accusative/possessed); adjectives, to agree with their nouns' definiteness, number, and gender (masculine/feminine); verbs, for tense or aspect (past/present/future or imperfect/perfect) and mood (indicative/jussive/imperative), and to agree with their subjects' gender, number, and person (first/second/third); and pronouns, for gender and number, and in the case of personal pronouns, person and case. I really should re-emphasize, though, that this varies from language to language within the Semitic family, and sometimes it depends whom you ask. Ruakh 04:33, 3 March 2006 (UTC)

[edit] Basque is polysynthetic???

I know it's not really relevant, but Basque is NOT a polysynthetic language. I'm going to edit the description of Basque as "an extremmely inflected polysynthetic language" to remove the word "polysynthetic," if noone minds.67.170.176.203 14:25, 10 April 2006 (UTC)

It's not incorporating, but it is polysynthetic. Ruakh 16:26, 10 April 2006 (UTC)

[edit] Germanic languages vs. Romance languages

On what is based the following comparison between Germanic languages and Romance languages?

"Some branches of Indo-European (e.g. the Slavic languages and the Germanic languages with the exception of English and Afrikaans) seem to have generally retained more inflection than others (e.g. Romance languages)."

I think it's difficult to make a comparison. While noun inflection is much simpler in most Romance languages (which have lost declensions) than in Germanic languages (some of which still have a few active declensions), with verb conjugation it's the other way around: Germanic verbs typically have much less synthetic tenses than Romance verbs. FilipeS 16:06, 22 April 2006 (UTC)

Not to mention that the statement is contradicted a few paragraphs below:

"The Romance languages like Spanish, Italian, French, Romanian etc., are more inflectional than English, especially when it comes to verb forms. A single morpheme usually carries information about person, number, tense, aspect and mood, and the verb paradigm may be considerably complex."

By the way:

"There is no Ablaut or Umlaut, and only little predictable vowel alternation, found on certain verbs where the Latin root had the phonemes /E/ or /O/."

Isn't ablaut just another name for vowel alternation? FilipeS 16:10, 22 April 2006 (UTC)

[edit] Newly added image.

Is the newly added image accurate? The masculine/feminine distinction looks to me like derivation rather than inflection. Also, the image technically depicts Catalan inflection, though I guess it happens to be the same in this case. —RuakhTALK 03:47, 3 January 2007 (UTC)

According to the definitions in the article, I would say it's inflectional. I like the picture. :-) FilipeS 18:08, 5 January 2007 (UTC)
What gives you that impression? The article doesn't give actual definitions so far as I can see, but it gives two loose criteria: first, whether the information it supplies is grammatical in nature, and second, whether a dictionary might list it separately. The sex of a cat doesn't strike me as being a bit of grammatical information (though it does have grammatical consequences, since Catalan has gender agreement for adjectives and determiners, and the nouns gato and gata are of different genders); and I don't know about Catalan dictionaries, but the Real Academía Española does list the nouns gato and gata separately. (Of course, the nouns have the additional complication that the noun gata has other meanings besides just "female cat", but the noun chica has an entry even though it says simply to see the entry for chico, so the RAE obviously does think that in sex-paired nouns, each noun warrants a headword). —RuakhTALK 06:09, 6 January 2007 (UTC)

What gives me that impression? This, for one thing:

Inflection is the process of adding inflectional morphemes (atomic meaning units) to a word, which may indicate grammatical information (for example, case, number, person, gender or word class, mood, tense, or aspect). Compare with derivational morphemes, which create a new word from an existing word, sometimes by simply changing grammatical category (for example, changing a noun to a verb).

FilipeS 16:25, 6 January 2007 (UTC)

Yes, I read that. But — as I said — the difference between the nouns gato and gata is not one of gender; it's one of sex. Eso animal es una gata ("That animal [←masculine] is a female cat") is a perfectly grammatical statement in Spanish. Now, it's true that the nouns do have different genders, one being masculine and one being feminine, but that's a consequence of the main difference between the nouns — the sex of their referents — and of Spanish's tendency to use natural gender. —RuakhTALK 19:32, 6 January 2007 (UTC)

You are making a needless complication with sex and gender, there. They are not totally disjoint things. That's a misconception that people whose native language does not have grammatical gender often have. The two genders are called "masculine" and "feminine" for a reason: the former is where you usually find nouns that designate males, and the latter is where you usually find nouns that refer to females. It's no different from singular vs. plural in grammatical number. Just as "car" and "cars" are two forms of the same lexeme in English, so are "gato", "gata" "gatos" and "gatas" in Spanish or Catalan. FilipeS 19:45, 6 January 2007 (UTC)

First of all, I have two native languages, and one of them (Hebrew) does have grammatical gender. I'm not saying that sex and gender are "totally disjoint", but they are slightly different, and this is exactly the context in which that difference is relevant: inflection has to do with grammatical information such as gender, not biological information such as sex. It's like how grammatical number doesn't necessarily have to do with actual real-world number (consider e.g. "many a man has", "more than one man has", etc.), but in the case of gender we have the convenience of having separate words for grammatical gender and natural gender ("sex"). By the way, you seem to be arguing somewhat of a straw man here, as I do accept that grammatical inflection exists; obviously the adjective gato(s)/gata(s) (meaning blue- or green-eyed) is one lexeme in Spanish (and in Catalan, if it exists in Catalan). I just don't believe that nouns are inflected for gender, else we'd have forms like computadoro (Esto objeto es un computadoro). —RuakhTALK 00:33, 7 January 2007 (UTC)

Then I guess nouns aren't inflected for number, either, else we'd have forms like "rices". And I guess verbs aren't conjugated in English, since "may" and "can" have no gerunds.

In my opinion, you are falling prey to a misconception, which is the notion that, if a language has inflection, then every word in it must exist in all possible "versions". That is definitely not the case. In the Romance languages, there are plenty of examples of defective verbs; that is, verbs which only exist in some of their theoretically possible inflectional forms. This happens either for semantic reasons (a particular form does not "make sense"), or for euphonic reasons (that form would sound funny, so a different verb altogether is used insted). But I don't think this means that the different forms of defective verbs are derivational, rather than inflectional. They are simply words whose inflectional scheme is incomplete.

Moreover, describing "gata" as derivational gives the impression that the word "gato" appeared first, and only much later was the feminine version invented, as a sort of afterthought. As a speaker of Portuguese, which is closely related to Spanish, that's not how I perceive "gato" and "gata" at all! The latter is not an "addition" or an "appendix" to the former; they are two versions of essentially the same thing.

"Sex" is a property of living beings, not of abstractions. It's not correct to apply it to nouns. Cats may be male or female, but the words "cat", "gata" and "gato" are neither. "Gata" is a feminine grammatical gender word, not a "female" word, and it is just as feminine grammatical gender as "mesa" or "buena". It is artificial to split the three into two classes on semantic grounds, because grammatical gender isn't primarily a matter of semantics in the first place. FilipeS 18:04, 7 January 2007 (UTC)

Pink and blue for female and male cats? Give me a break. It may be fairly illustrative (though a bit too cute to be taken seriously), but there are some unfortunate implications in such a picture. Please try to figure out a solution that is a tad more neutral when it comes to perpetuating traditional gender roles.
Peter Isotalo 19:30, 20 February 2007 (UTC)
Retrieved from "http://en.wikipedia.org../../../i/n/f/Talk%7EInflection_e780.html"