Talk:Infinitive

Archive

• Archive1 — discussions from 2 January 2002 to 11 July 2005

Rewrite of intro?

Currently, the lead text is rather vague, and yet manages to contradict itself. I think the problem is that the term infinitive is used differently in discussing different languages. For example, most languages AFAIK have only impersonal infinitives, but Portuguese (apparently? [Indeed! Velho]) has personal infinitives; in most languages AFAIK, infinitives are verbs' "dictionary forms," but some languages have no infinitives, and Hebrew has them but uses the third-person masculine singular past indicative for dictionaries; in most languages AFAIK, an infinitive cannot be cannot be a "main verb," but French allows the infinitive to double as a sort of abstract imperative in signs/postings; and so on.

I don't think there's a universal definition of the term that applies to its (commonly accepted) uses in all languages; therefore, I think the solution is to give a first sentence that explains that the term describes similar but distinct forms in different languages and introduces a list of features of infinitives that are common among many language (together with notes on deviations from these commonalities).

It's probably also worth noting - though perhaps not in the lead text - when non-infinitive verb forms share these features; for example, English has a "gerund" form that acts the same as many other languages' infinitives, and that is indeed interchangeable (or almost interchangeable) with English's infinitive in some contexts (e.g., "I saw him run" vs. "I saw him running").

Does anyone have any thoughts on the matter? Ruakh 07:40, 4 September 2005 (UTC)

I agree with the criticism. What characterizes the personal infinitive is that it is inflected, meaning that it has different forms according to person and number. So I cannot agree with the definition currently in the article:

"In grammar, the infinitive is the form of a verb that has no inflection to indicate person, number, mood or tense."

The infinitive can be inflected to indicate person and number, in some languages. What it does not show is tense -- the time at which an action occurs --, or mood. FilipeS 23:08, 10 May 2006 (UTC)

Non-sense: infinitives may inflect for tense: "It's good to write a book!" / "It's good to have written a book!" Or, e.g. in Portuguese, "É bom escrever um livro!" / "É bom ter escrito um livro!". Velho 00:07, 11 May 2006 (UTC)

The infinitive forms in those examples, to write, to have, escrever, and ter, are not inflected. You are confusing compound verb phrases with conjugations. FilipeS 14:24, 11 May 2006 (UTC)

1. Oh dear! And I thought that there were compound verb forms! Ts, ts, ts... But now, how many tenses are there in English, for instance?

2. Notice also that the auxiliary may be inflected for person in languages that allow it: "Foi bom teres escrito esse livro!" (It was good that you wrote that book).

3. Are you sure you wanted to say "verb phrases"? Verb phrases include the complements and the adjuncts of the verb.

See you around! Velho 18:06, 11 May 2006 (UTC)

As you know, modern linguists tend not to count compound verbs as tenses (you've probably heard that "English has no future tense" at some time in your life, for example), since they are not constructed by inflecting the main verb. Otherwise, there would be no difference between synthetic languages and analytic languages, as far as verbs are concerned. FilipeS 18:46, 11 May 2006 (UTC)

The difference between s. languages and is a. languages is quantitative. I would almost agree with you if you said "some" instead of "modern". Check Wikipedia pages on languages' grammars: compound verb forms are included in... verb forms. So, an infinitive may be compound. It is still an infinitive, whether it is an inflected form or not. (comment continues below)

It is still an infinitive, whether it is an inflected form or not.

Did I ever claim that inflected infinitives were not infinitves? Kindly read more carefully, the next time. FilipeS 14:47, 12 May 2006 (UTC)

You've probably heard that at some time in your life that there are a few differences between compound tenses and peryphrastic structures. For instance, auxiliary verbs in compound tenses do not select any arguments. Verb complements and adjuncts cannot be inside the compound (between auxiliary and compound).
Tell me: what is an auxiliary verb?
By the way, have you ever noticed that Latin has a synthetic perfect infinitive? E.g., amāvisse, fecisse, etc.
See you! Velho 23:17, 11 May 2006 (UTC)

You've probably heard that at some time in your life that there are a few differences between compound tenses and peryphrastic structures.

There are no semantic differences, but there are morphological differences.

By the way, have you ever noticed that Latin has a synthetic perfect infinitive? E.g., amāvisse, fecisse, etc.

So what? I never said the infinitive couldn't be inflected for aspect.

Here's the deal: the definition currently in the article is wrong. It's wrong if we only count inflected verb forms, as I was doing, and it's wrong if we take your approach, too. How do you propose to correct it? FilipeS 14:52, 12 May 2006 (UTC)

I suppose the synthetic Latin forms are in fact preterite infinitives. They're used either for perfect or imperfect clauses. But the point is different: regardless the languages that you and I know, there is nothing forbiding that, in some languages, an infinitive be inflected for tense.

The difference between compound and peryphrastic structures may be semantical. If the semi-auxiliary in a peryphrastic structure selects arguments, then we have a semantical difference.

About the deal: you may have noticed that we absolutely agree that the definition is wrong. I'm not sure about the best way to fix it, or I would have done it already:

1. I guess that there is no way to define "infinitive" for every language without including, e.g., the English gerund or the Latin supine.
2. However, there should be made a difference between infinitives and (other) verbal nouns, but I don't think that this difference may be kept in every language.
3. So, we must distinguish a universal definition and a definition that works for languages that have two infinitives according to the previous definition, but only call "infinitive" one of them.
4. The universal definition must distinguish infinitives from other non-finite verbs (notice that Wikipedia's definition of non-finite verb is wrong too).
5. I guess that infinitives, in the "universal" way of seeing them, should be defined more or less this way: they function as verbs within the infinitive clause (or predicate), but they make this clause (or predicate) function as a noun.
6. The finite uses of infinitives, as in some French and Portuguese imperative sentences, should be seen as "exceptional", meaning that those infinitives are called "infinitives" only for morphological (or etymological...) reasons, but they aren't "real" infinitives.
7. If you agree with this, perhaps we could do something about the definition...

Please stop indenting the text, or this will soon be unreadable! :-)

Velho 15:45, 12 May 2006 (UTC)

Re: "Please stop indenting the text, or this will soon be unreadable!": Sorry, that was me; I didn't think you'd mind. I think the unindented version is unreadable — it's easier for the two of you, because each of you knows what he added and what the other guy added, but for the rest of us, there's just a huge blob of text. If you could find some way to make it clearer what is responding to what, whose comments end where, etc., I'd appreciate it. Ruakh 14:10, 13 May 2006 (UTC)

Still about tense: in some languages, finite verbs do not inflect for tense, even if they inflect for aspect or, sometimes, person. That's the case with Portuguese Creoles. Velho 16:29, 12 May 2006 (UTC)

Then what makes them "finite"?...

BTW, these latest indentations were not my doing. FilipeS 16:41, 12 May 2006 (UTC)

They are used in finite sentences, no? And they "inflect" for aspect (with the help of a particle/prefix). They are also the head of simple declarative, imperative, etc., sentences. Velho 17:12, 12 May 2006 (UTC)

Yeah, but any inflection of a verb can be used in a finite sentence. At least, from my limited experience with languages.

As for being inflected for aspect, I think it can be argued that the same happens with the non-finite forms of Portuguese and English, for example. FilipeS 18:18, 12 May 2006 (UTC)

Hey, shouldn't you say something about my "proposals" for a definition of infinitive? Velho 18:55, 12 May 2006 (UTC)

I hadn't realised there were any. Where are they? FilipeS 19:15, 12 May 2006 (UTC)

Wait, I get it now:

"In grammar, the infinitive is the form of a verb that has no inflection to indicate person, number, mood or tense. It is called the "infinitive" because the verb is usually not made "finite", or limited by inflection. In some languages, however, there are inflected forms of the infinitive denoting attributes such as tense, person and number. For example, Portuguese inflects infinitives for person and number; Latin inflects infinitives for tense. In several languages, including English, compound infinitive forms are different for present and perfect (e.g., "to sneeze" and "to have sneezed"). On the other side, some languages do not have infinitives at all, for example Arabic, Bulgarian and Modern Greek. In some languages the infinitive can be construed as a verbal noun."

"On the other hand" is better. It is clearer now, although there is still a contradiction with the first sentence:

• Infinitives are not supposed to be inflected for person, number, mood or tense, yet some are.
• If not being inflected for person, number, mood or tense is all there is to being an infinitive, then gerunds and participles should count as infinitives, too -- which we don't "want"...

Here's a question: why did the Romans call it the "infinitive"? FilipeS 21:26, 12 May 2006 (UTC)

Is it possible that there simply is no general definition of an infinitive? I mean, I'm pretty sure that if English were some newly-discovered language that linguists were first characterizing, they'd label our gerund as a kind of infinitive. And with Portuguese, I'm pretty sure they wouldn't label its personal infinitive as an infinitive at all. I stand by my original proposal, which is to explain that the word infinitive means different things in different languages, give some features of infinitives that are common in most languages, and discuss variations on such. Ruakh 21:49, 12 May 2006 (UTC)

And perhaps different things for linguists and in traditional grammar... FilipeS 22:30, 12 May 2006 (UTC)

Come on, Filipe. I was talking about the sentences above numbered from 1 to 7.

The Latin perfect infinitive and the Portuguese personal infinitive are infinitives because they cannot stand alone in a simple sentence, the clause they're the head of functions as a noun within the main clause, and the infinitives themselves can be treated more or less as nouns (although they function as verbs inside the infinitive clause).

Velho 13:11, 13 May 2006 (UTC)

There is a reason not to call the English gerung an "infinitive": it is an infinitive only sometimes.

"Eating good food is great." This is syntactically speaking an infinitive. The embedded clause ("Eating good food") works as a noun (or a NP).

"She's a sleeping beauty." Syntactically, this is not an infinitive. "Sleeping" functions as an adjective.

"I broke it throwing it against the wall." This is not an infinitive either. "Throwing" functions as an adverb.

Velho 13:19, 13 May 2006 (UTC)

Of your examples, only the first uses a gerund; in your second and third examples, sleeping and throwing are participles. Further, the English bare infinitive cannot serve as a noun (*"Eat good food is great"), while the to-infinitive, which can serve as a noun, can also serve as an adjective or adverb: "This food is to die for (i.e. worth dying for)", "I have so much to do (i.e. that I need to do or that I can do)", "I went there to see what I could see (i.e. in order to see what I could see)". So, by traditional definitions, the gerund is more consistently a noun than the infinitive is. Ruakh 14:22, 13 May 2006 (UTC)

About indenting and readability: Ruakh, I think indents are ok, but they may go up and down, and not necessarily always up! I also suppose that the text of the section will be more readable if everybody always writes at the end of it. Velho 14:28, 13 May 2006 (UTC)

About gerund: you're obviously right. Sorry. About the to-infinitive: hmm, hmm, hmm... You're probably right here too. Anyway, the noun thing is not a traditional definition. Velho 14:28, 13 May 2006 (UTC)

And then again... Are you sure about the third example ("I broke it throwing it against the wall.")? Is it a participle? Indeed, I guess gerund was named "gerund" because the Latin equivalent to these sentences is made with a gerund or a gerundive (as amando or amandus/a/um). Velho 16:38, 13 May 2006 (UTC)

Re: Only adding comments to the end: yeah, that should help.

Re: The name gerund: It definitely does come from Latin; indeed, the OED defines gerund as "A form of the Lat. vb. capable of being construed as a n., but retaining the regimen of the vb. Hence applied to forms functionally equivalent in other langs., e.g. to the Eng. verbal noun in -ing when used rather as a part of the vb. than as a n." That said, it seems to be well accepted that a gerund is always a noun. (That said, it also seems to be well accepted that a participle is always an adjective; in your throwing example, it's construed as an adjective modifying I. That's why people speak of dangling participles, which are participles that are separated from the nouns they're supposed to modify to the point that the noun they seem to modify isn't the noun they're supposed to. I think it makes more sense to take the participle as an adverb in this case, as Spanish does, but *shrug*.) Incidentally, the OED's first citation for gerund is the quote, "There be moreouer belongyng to the infinitiue mode of verbes certayn voyces called gerundes..whiche haue bothe the actyue and passiue significacion," which is a different way of looking at it from ours. (Not that a quote from 1513 is really relevant to the modern sense of the word, but I thought it might interest you.)

Ruakh 17:31, 13 May 2006 (UTC)

My thoughts on your reflections above, Velho...

It seems that there are two different concepts of "infinitive" (at least):

• One is the old notion of infinitive, in the sense that it has been traditionally applied -- and still is -- to ancient Greek and ancient Latin, and to other Indo-European languages, but which may well not be relevant to other language families.
• The other is a modern, syntactic definition which is general enough to be applied to any language, but won't necessarily coincide with the old concept of "infinitive" in all cases.

Why not discuss both in the article?

Another thing is that, particularly judging from your comment nr. 6, you are regarding the "infinitive" as an essentially syntactic concept ("It's part of a noun clause, ergo it's an infinitive"). That may be a useful perspective for those who study language from a linguistic point of view, but it seems to me that language learners are likely to be more interested in seeing the infinitive as a morphological category ("How many different forms of this verb-thingy do I have to memorize, and which is the simplest?") FilipeS 18:08, 13 May 2006 (UTC)

Well, we're not far from complete consensus... I vote for two definitions, one of them morphological, language dependent and aimed at "language learners"; the other one syntactical and more useful "from a linguistic point of view"... You (Ruakh and Filipe) could go ahead and write them out, couldn't you?... ;-) Velho 23:44, 13 May 2006 (UTC)

Honestly, after our conversation, I don't think I'm qualified to do that. :o FilipeS 16:32, 15 May 2006 (UTC)

Proposed wording for lead text.

Here's my proposed lead text:

In grammar, the infinitive is a non-finite verb that exists in many languages; in English, the infinitive of a verb is its basic form with or without the preposition to: so, do and to do, be and to be, and so on are infinitives. As with many linguistic concepts, there is not a single definition of infinitive that applies to all languages; however, in languages that have infinitives, they generally have most of the following properties:

• They head verb phrases which can contain objects, adverbs, predicatives, and so on.
• They function as other lexical categories — usually nouns — within the clauses that contain them, for example by serving as the subject of another verb.
• They do not represent any of the verb's arguments (as employer and employee do).
• They are not inflected to agree with any subject, and their subject, if they have one, is not case-marked as such.
• They cannot serve as the only verb of a sentence.
• They are the verb's lemma or citation form (the name of the verb, often regarded as its basic uninflected form), used in giving the definition or conjugation of the verb.
• They do not have tense, mood, aspect, and/or voice, or they are limited in the range of tenses, moods, aspects, and/or voices that they can use.
• They are used with auxiliary verbs.

However, it bears repeating that none of the above is a defining quality of the infinitive; infinitives do not have all these properties in every language, and other verb forms may have one or more of them. (For example, English's gerunds and participles have most of these properties as well.)

Then I figured the sections on individual languages could mention the ways they deviate from this list.

What do y'all think?

Ruakh 02:09, 16 May 2006 (UTC)

I think it is a very good idea and it is very well done. Perhaps the wording of the first paragraph could be better, and one or two of the "properties" might be changed (especially the first one), but I think Ruakh's text could be inserted in the article right now. Velho 02:46, 16 May 2006 (UTC)

The first "property" could perhaps be "Infinitives head verb phrases which, as any verb phrases, can contain objects, adverbs, predicatives, and so on, and be c-commanded by a subject." Velho 02:51, 16 May 2006 (UTC)

Hmm. It seems to me that in English, the c-commanding can go the other way: in for him to do it, it seems that to do it c-commands for him and hence him, but him only c-commands for. And in French, in something like J'ai demandé à Jean de le faire (literally I{{'ve asked} {to Jean} {of {it {to do}}}}, really I asked Jean to do it), it seems that neither of Jean and faire c-commands the other, as each is buried within its own prepositional phrase. Now granted, your wording only says that an infinitive can be c-commanded by a subject, but to me that sounds like "An infinitive can have a subject, in which case the subject c-commands it," even if what you meant was "An infinitive can have a subject, in which case the subject may c-command it." Ruakh 12:58, 16 May 2006 (UTC)

Well... I just think that we should say that an infinitive can have a subject (as much as an object or an adjunct). Another thing: clearly, infinitives are non-finite verbs, but the current definition of "non-finite verb" is clearly wrong, especially because ti says that non-finite verbs cannot be "limited", which is wrong in every sense. It also says that non-finite verbs cannot work as predicates, which is even worse. Velho 18:04, 16 May 2006 (UTC)

It seems like the right direction to take, but I have a few comments:

They cannot serve as the only verb of a sentence.

Tone this down a little bit. Quite a few languages use infinitives as imperatives. (I know that according to Velho that makes them not-real-infinitives, but still...) I suggest "They do not serve as the only verb of a sentence."

They are the verb's lemma or citation form (the name of the verb, often regarded as its basic uninflected form), used in giving the definition or conjugation of the verb.

Tone this down, as well. Infinitive = name of the verb is actually a very modern-Western-European notion. In classical Greek and classical Latin, for example, verbs were named by the first person of the present tense (amo, canto, etc.) Besides, some languages have no infinitive at all.

They do not have tense, mood, aspect, and/or voice, or they are limited in the range of tenses, moods, aspects, and/or voices that they can use.

How are you going to distinguish infinitives from gerunds and participles?... FilipeS 18:47, 16 May 2006 (UTC)

Replying to both Velho and FilipeS at once:

Re: "I just think that we should say that an infinitive can have a subject [...]": my text already does say that: "They are not inflected to agree with any subject, and their subject, if they have one, is not case-marked as such."

Re: the problems with Non-finite verb: I agree, and that really must be fixed, but I don't see that that affects this article.

Re: languages that use infinitives as imperatives: How about, "They cannot serve as the only verb of a declarative sentence"?

Re: the infinitive being the name of the verb: Well, how common is it? If >50% of languages with infinitives use them as lemmata, then given all the disclaimers surrounding the list, I think it's valid to include it without comment. If not, then I suppose it can be tacked on afterward in its own paragraph or something.

Re: Distinguishing from gerunds and participles: Unless someone can find a source that explains the difference cross-linguistically, I don't think we can really draw that distinction ourselves.

Ruakh 20:10, 16 May 2006 (UTC)

A suggestion for the first paragraph and the first property:

In grammar, infinitive is the name for certain verb forms that exist in many languages. In English, the infinitive of a verb is its basic form with or without the preposition to: so, do and to do, be and to be, and so on are infinitives. As with many linguistic concepts, there is not a single definition of infinitive that applies to all languages; however, in languages that have infinitives, they generally have most of the following properties:

• In most of their uses, infinitives are non-finite verbs.

My idea is to stress that, within each language, infinitives are defined firstly by a morphological criterion, that is, only after we identify a morphological category in verbs of a certain language will we call it "infinitive". And we will do it if it has some or most of Ruakh's properties. I suppose that we could even write "...name for certain morphological verb forms...". Velho 23:20, 16 May 2006 (UTC)

Okay, since we seem to be agreed that this text is better than what was there before, I've instated it (after incorporating some of the changes the two of you suggested). Further changes are of course welcome; I just figured that since the previous text was obviously bad, we might as well replace it as soon as possible and make further improvements afterward. Ruakh 03:15, 17 May 2006 (UTC)

It's great! Now I think it could be useful to have a first section (after introduction) called "Infinitives and their definition across languages" stating how each of the general "properties" can not be applied in some languages. I'll do it today or tomorrow. Velho 18:40, 17 May 2006 (UTC)

I look forward to it. :-) Ruakh 20:56, 17 May 2006 (UTC)

Alternate for INFINITIVE

Infinitive is also the name of a management consulting company based in the DC Metropolitan area (US). www.infinitive.com —The preceding unsigned comment was added by 209.183.210.226 (talk) 18:58, 14 May 2007 (UTC).

Infinitve inflection in NIA languages

"These are the only Indo-European languages that allow infinitives to take person and number endings. " This is not true. in NIA (New-Indo-Arian) such as Hindi, the infinitive can be inflected for gender, case etc. As in "jaane ke liye" (from the infinitive Jaana - To Go, Jaane - oblique case) or "meri jaani ho cuka hai" (Jaani for female gender). —The preceding unsigned comment was added by 89.1.118.212 (talk • contribs) 10:35, 16 June 2007 (UTC).

Dare as the main verb

This seems to be able to take both the bare and full infinitive. To my mind anyway, both of these seem grammatically correct and mean the same thing:

He dares oppose us. He dares to oppose us.

Now, my source for the direct quote of the first is from C&C3: Kane's Wrath (level 2 or 3 I think), so I don't know if the writers used bad grammer, but it was definitely "He dares oppose" and not "He dare oppose" as the article on modal auxillary verbs suggests.