Talk:Predicate (grammar)

Change her to him

The last edit changed it to "she" was elected president. Obviously someone salty over donald trump winning. we should ban this person.

Adjectives and nouns are predicates?

Can someone explain what the latter half of this sentence means?

This notion is closely related to the notion of a predicate in formal logic, which includes more expressions than the former one, like, for example, nouns and some kinds of adjectives.

I find this sentence extremely confusing. I don't see how an adjective or noun can be a predicate.--Subversive Sound (talk) 00:49, 13 March 2010 (UTC)

I'd like to see some examples too. Seems like in linguistics, when someone asks, "what color is that cat." and another responds with just "yellow", there response acts as predicate (lacking a subject, and lacking a copula). Someone who studies logic should elaborate. As is, the sentence is pretty useless (and may not even be valid).24.116.83.122 (talk) 21:49, 12 November 2010 (UTC)

Where do I begin with criticisms. First, let me sat that most of the examples cited by the author are constructed with very poor grammar. ' Whom did Jim give his dog to? ' let's try it in formal english ... 'To whom did Jim give his dog.' and another ... — Preceding unsigned comment added by 50.72.100.79 (talk) 04:04, 26 February 2013 (UTC)

I have corrected the example from Whom... to Who.... Your example, To whom..., is stilted English. Yes, you may have learned that one should not strand a preposition, but that is prescriptive grammar, which in this case this author views quite negatively (because few people actually speak like that). Any other comments?

disambiguation requested

I agree that the term 'traditional grammar' as used in this article is gratuitous, confusing, and perhaps intentionally dismissive. All the stuff about linguistic semantics should be taken out and put into a separate article. Furthermore, the notion that a predicate describes what may be true of the subject is a definition fraught with cultural and political minefields that are completely unnecessary to understanding English grammar. If I say 'Black people are evil.', is that therefore 'true'?! And if it isn't 'potentially true', is it therefore not a predicate? If not, then what is it? All of these issues are avoided when the predicate is simply what the speaker/writer says about the subject, and the issue of 'truth' is left out. Natcolley (talk) 15:59, 11 April 2009 (UTC)

biased English POV?

What about Chinese - it doesn't need a copula to equate a subject to an adjectival predicate. French has many of its object pronouns come before the verb. Marking NPOV and world on this one. John Riemann Soong 01:25, 16 September 2006 (UTC)

Last I checked, this is the english version of wikipedia, and thus this content is relevant to the english language. You may wish to expand on the article to include grammar in other languages, but this is hardly a problem with point of view. Removing flag... Tiranak

Last time I checked, the notion of "predicate" was relevant to more languages than English. I do agree that it's natural for the English wiki to exemplify things in that langueage, though, so I agree that the flag should be removed. But take it easy. Neither 05:15, 21 October 2006 (UTC)

The language of the site is no excuse for biased coverage. By extension of this tendency, the article on eye should be exclusively about the human eye on every language wiki, because this is the human Wikipedia. We should also not cover the subjunctive mood outside of the English case, nor non-English grammar in any case where the grammatical construct exists in English. ᛭ LokiClock (talk) 08:26, 22 August 2010 (UTC)

Okay, that was a stupid argument. Let me try again. Your argument, Tiranak, was basically that because it's an English language article that an English bias should be permitted. But this rests on the idea that the article is about English because it's in English and about languages, one of which is English. If this were Predicate (English grammar), it would be another thing. Remember also that people will run into the term while learning a foreign language, and that an English-biased article will be completely unhelpful to them. ᛭ LokiClock (talk) 19:50, 24 September 2010 (UTC)

I strongly welcome the inclusion of more information on Chinese predicate semantics. The content should probably be added at the end of the English section in a subheader titled "Predicates in Non-English Languages" , not to show any kind of bias against Chinese grammar, only to first explain what a predicate is in the host language. This structure simplifies the matter of elucidating predicates, with out having to add a ton of exceptions to the rules of what predicates are in the opener. Who knows, if we get enough content here, we might be able to split the article into separate pages where each topic can be in the "spot light" to the reader. I don't know enough about French or Chinese to add content to this page, but you shouldn't think that it's being restricted or banned, it just happens that very few people know how non-English predicates work.Xetxo (talk) 22:14, 12 November 2010 (UTC)

I agree with those points on the order (English before others), but I believe "Predicates in other languages" is a title that better encourages languages to outgrow the subsection into one of their own. ᛭ LokiClock (talk) 10:22, 13 November 2010 (UTC)

Predicates in logic and linguistics

I think that this article conflates the different ideas of predicates in logic and linguistics. I think it should be clear which is which. In traditional European grammar, as the article says, the predicate is the part of the sentence following the subject (ie verb, objects and adjuncts). Later, in logic, it came to be used for the class of units that in languages roughly correspond to verbs and adjectives. In Modern linguistics the word was borrowed back again from logic, and used by people like Vendler to refer to verbs (*not* including objects). So, I think the first sentence of the article is misleading. Linguistics, per se, does not have the notion of propositional truth. The statement that a predicate is something that has a truth value, could be dropped altogether. If it is retained, I suggest it is introduced with something like "On the other hand, in logic ...". Secondly, I think the statement is factually incorrect. It is a proposition, rather than the predicate that it contains, that has a truth value. Here's a supporting quote from R. L. Trask, 1993, "A Dictionary of Grammatical Terms in Linguistics":

"1. That constituent of a sentence, most typically a verb phrase, which combines with the subject NP to make up the complete sentence. ... 2. A verb, or a complex structure consisting of a verb or auxiliary plus a closely bound meaningful element, when this is considered as linguistic unit which can or must combine with specified arguments or participant roles to make up a clause. Grammatically, they are classified in terms of their valency, the number and types of uints which they require. There are also grammatically relevant semantic classifications, the best-known of which is that of Vendler (1967). ... 3. In formal logic, an element which must combine with a specified number of arguments to make up a well-formed expression; the linguistic use of 'predicate' in sense 2 is directly derived from this logical usage."

This is my first time contributing to Wikipedia, so please excuse me if my wiki-ettiquette is lacking. Brian.Murphy@cs.tcd.ie

I agree that there is a discrepancy between the notion of a predicate in trad grammar and that in current linguistic theory, but I disagree that there's any conflation in the article as it was. The point is that current semantic theory is a branch of applied mathemetical logic. And this is not just a parochial view, but rather the mainstream approach to the meaning of natural language these days. Also the statement in the first paragraph is not that a predicate has a truth value but rather, that it is true of something. This is perfectly in keeping with practices in both logic and linguistics, and it's not controversial. But the previous version did downplay the traditional notion by not mentioning it in the first paragraph, and I agree that that needed improvement. So I've tried to remedy that in the new version. Concerning the Vendler classes, it's true that Vendler himself used them to classify verbs, not predicates, but there is consensus now that the classification applies to predicates rather than verbs. In fact, the paragraph about the Vendler classes discusses this, and there are references below to back it up. I've also removed the flag. Neither 13:48, 20 October 2006 (UTC)

You have missed a crucial concept; Mathematics, and formal Logic also, are both languages. Mathematics is a language, or several languages, one might argue. As a language(s), Mathematics suffers from most of the geopolitical ills of the world, and is often poorly spoken and even more poorly taught in schools. — Preceding unsigned comment added by 50.72.100.79 (talk) 04:38, 26 February 2013 (UTC)

Every example in this article has the verb in either the present or past tense. As a layman who knows nothing about formal linguistics, it strikes me that lots of the things mentioned here change with future-tense verbs. E.g.
John will be sick
cannot be a static Vendler class the way "John is sick" is as we can say
John will be sick in an hour
(he just ate a bad oyster). A few examples from you linguistic types would help clear up some confusion amongst us non-linguistic types.--Longpete 08:46, 4 February 2007 (UTC)

yeah, you're right, things change with tenses. But what happens when you use "in an hour" with future tense is that it takes on a slightly different meaning from what it normally has, something like, ``an hour from now." That seems to be a property of the expression "in an hour" in English. Other languages have separate expressions for those meanings of "in an hour." The Vendler classes are supposed to apply to any language, so one has to be careful what tense one puts the predicate in. Future is a bad choice, since it will influence the result, as we have just seen, and the present progressive (John is running) is another bad choice. The best choice of tense to carry out the in an hour test in English is simple past. That doesn't show that the test is wong, though! Neither 04:25, 16 March 2007 (UTC)

Shoot, I just made an edit to the intro paragraph after reading the discussion for another page instead of this one. I'll post here what I posted in the mistaken discussion page under the header "definition intro" although the topic may relate closely to this header.Xetxo (talk) 20:03, 23 April 2009 (UTC)

s-l versus i-l predicates, and k-l predicates

I don't follow the argument here.

"John is hungry" is given as an example of an s-l predicate. However, in the section on i-l predicates, we have:

There are firemen available. ("available" is s-l)

• There are firemen altruistic. ("altruistic" is i-l)

This leads me to conclude that "There are firemen hungry" is well-formed, which it isn't. Of course, we can say "There are hungry firemen", but we can also say "There are altruistic firemen".

It isn't clear to me in these "There are..." sentences what the subject is - 'there'? --Nyelvmark 19:38, 8 June 2007 (UTC)

Nope, in a sentence like "There are firemen available." the subject is "firemen" and "there" is an adverb. This is a peculiar construction, to be sure, but very common in English. 24.27.31.170 (talk) 18:26, 11 June 2011 (UTC) Eric

As one example "police" is gender neutral. and the other "firemen" is gender specific, wouldn't it be a clearer example if either gender neutral or gender specific were used in both examples? —Preceding unsigned comment added by 82.22.127.197 (talk) 22:44, 3 October 2009 (UTC)

About "kind-level" predicates, I think a common example of misuse is the word 'key' and in "Our business partners are key to our success." Why people say this is a mystery to me -- you could say "Our business partners are the key to our success." and it would be perfectly good grammar. 24.27.31.170 (talk) 18:26, 11 June 2011 (UTC) Eric

I wouldn't say "There are firemen hungry", but on the other hand, as a child growing up in the 1950s, my parents occasionally repeated what their parents said to them, when we didn't finish the food we'd been served: "There are children starving in China." So perhaps "There are firemen altruistic in China" is okay. 208.54.85.172 (talk) 07:02, 10 May 2014 (UTC)

I think you don't understand how to parse those sentences. In the sentence "There are children starving in China," the verb is the progressive aspect of the verb "to starve" (i.e., "are starving"). In the sentence "There are firemen altruistic in China," the main verb is the present tense of "to be" followed by a predicate adjective ("altruistic"). The sentences are therefore very different grammatically; one is fine, while the other is atrocious. 2604:2000:EFC0:2:7C16:4E2E:E4E5:E5E8 (talk) 19:14, 5 August 2016 (UTC)

"In traditional grammar..."

Okay, "traditional" is being used often throughout the grammar wiki pages. Since "modern" grammar is not discussed, either we need to state "In traditional and modern grammar..." or drop "traditional".

I suggest dropping "In traditional grammar," and simply begin with "A predicate is...". Lets use the active voice.

What on earth is "modern grammar"? Grammar that violates traditional rules, perhaps? Also, the use of the the phrase "In traditional grammar" to begin the sentence has nothing to to do with voice -- the sentence uses the active voice with that phrase in place. 24.27.31.170 (talk) 18:29, 11 June 2011 (UTC) Eric

Traditional grammar is a tradition among speakers of a language of a particular grammatical analysis of their language, sometimes specifically traditions based on comparisons with Latin grammar. Traditional grammar does not contrast with "modern grammar," but with modern linguistic analyses of grammars. Hence the mention of verb phrase, which is a construct of modern linguistics. ᛭ LokiClock (talk) 07:32, 14 June 2011 (UTC)

Definition / intro

I'm seeing a possible split of the first paragraph into two, given that 2 definitions are being listed. The first definition also appears to need more beef. I've worked out the following and will post it here since my revision may need discussion and further revision:

In traditional grammar, a predicate is one of the two main parts of a sentence (the other being the subject, which the predicate modifies). For the simple sentence "John [is yellow]," John acts as the subject, and is yellow acts as the predicate, a subsequent description of the subject headed with a verb.

In current linguistic semantics, a predicate is an expression that can be true of something. Thus, the expressions "is yellow" or "is like broccoli" are true of those things that are yellow or like broccoli, respectively. This notion is closely related to the notion of a predicate in formal logic, which includes more expressions than the former one, like, for example, nouns and some kinds of adjectives.

The second definition is very unclear to me as it relates to "formal logic," I hope I'm not mistaken in thinking that "like broccoli" is only part of a predicate, which was the original error that brought me to this revision.Xetxo (talk) 20:07, 23 April 2009 (UTC)

Bold bits

There are some bold bits in the examples, which goes unexplained. "In the following examples, the predicate is underlined." - the bold parts seem to be subsets of the underlined sections, usually contain a noun, sometimes a preposition, but otherwise seem to have no rhyme nor reason. "me", "the book", "to the radio", "president", "in the park". What's being illustrated here? Needs clarification. 70.116.64.148 (talk) 17:45, 30 January 2010 (UTC)

Ascription

This page exhibits an extremely aggravating tendency in Wikipedia: useless redirections. I entered the search term "ascription", I was redirected here and there is no mention of ascription anywhere on the page. What is the point of a redirect if the term the searcher is looking for is not covered on the page he/she is directed to? Could someone who knows what they are talking about please at least cover "ascription", or take out the redirect, because at the moment the redirect is worse than useless. I now have to try to find another site that will tell me about "ascription", because Wikipedia can't. Zythophile (talk) 10:43, 28 March 2011 (UTC)

Zythophile, I recommend you post the unnecessary redirections to Wikipedia:Redirects for discussion. Thanks! --Michaelzeng7 (talk) 00:58, 7 December 2011 (UTC)

Dependency Grammar

I have nothing against using specific theoretical frameworks when that's particularly relevant, such as when different frameworks treat a phenomenon differently in important ways. I'm completely at a loss with regard to the relevance of dependency grammar to describe the notion of predicate, however. I suggest remove it and stick to the topic. :) Neither (talk) 19:27, 28 December 2012 (UTC)

What is not relevant about the use of dependency grammar? The article as it now stands makes mention of both phrase structure grammar and dependency grammar, considering the ways that both approaches understand predicates. What is not relevant about that? If there is something that is not clear in the article, what is it? --Tjo3ya (talk) 21:55, 28 December 2012 (UTC)

What is crucially missing in the discussion is the important consideration that it is the subject, and only the subject, that determines the person, gender, and number of the verb. Also, by redefining "predicate", it complicates a number of issues. For example, why is "leg" part of the predicate, when nobody would consider it part of a verb? As the object of a verb, it makes sense, but when predicate is redefined in this way, it seems to be an "argument"! In phrase structure grammar, the subject is easily recognizable as the noun or noun phrase to which the verb applies, and the rest of the sentence is the predicate. In dependency grammar, it isn't clear how the predicate differs from the verb, nor how one of the arguments (the subject) stands out as the pivotal topic about which the sentence revolves. (I'm disregarding sentences like "It's often been said that....") 172.56.26.198 (talk) 08:43, 10 May 2014 (UTC)

The understanding of predicates the comment is advocating is stuck in traditional grammar. It is stuck in a mode of construing predicates that does not get beyond the binary NP-VP division. In a sentence such as She is pulling his leg, the idiomatic expression is pulling...leg is the matrix predicate. The noun leg is part of this predicate. We know that it is part of the predicate because of the idiomatic meaning. She is not actually yanking on his leg, but rather she is teasing him by telling him something fictitious. The predicate in such a sentence is the catena is pulling...leg, whereby this predicate takes the two arguments she and his.

The basic problem with the traditional approach has to do with a paradox that it generates. If the predicate is everything but the subject, then there is no way to consistently distinguish between predicates and arguments. In a sentence such as He eats meat, the traditional analyis views eats meat as the predicate, yet this predicate contains one its arguments, i.e. meat. We have the paradoxical situation of the argument of a predicate being part of that predicate itself. The more modern approach to predicate-argument structures is not confronted with this problem, since it views just eats as the predicate, whereby this predicate takes the two arguments he and meat.

The subject in dependency grammar stands out by virtue of two facts. First, it is the argument that is ranked highest. It is ranked higher than first object, second object, and prepositional object; and second, it is always a dependent of the finite verb, which means it often appears higher in the syntactic structure than the object, since the object is always a dependent of the lexical full verb. To see this, have a look at the trees in the article. Look at where the subject is positioned in comparison to where the object is positioned. --Tjo3ya (talk) 17:44, 10 May 2014 (UTC)

Recent additions

The recent additions to the article by an unnamed user seem to have been performed by a knowledgeable person with good intent. Unfortunately, the additions are not coherently integrated into the entirety of the article. For instance, they include links to unrealized pages. I am now going to undo the revisions, but I encourage the person who performed the edits to register as a user and to respond here. The article can indeed be improved in the direction of the edits. --Tjo3ya (talk) 20:46, 30 November 2013 (UTC)

Purpose

"...the purpose of the predicate is to modify the subject." I don't believe that at all. I can say "She dances" without the least intention of affecting her ("She") in the least. Modification is out of the question. Whatever happened to the notion of a "verb", signifying "action" or "state of being"? 172.56.26.93 (talk) 06:46, 10 May 2014 (UTC)

Concrete reflex!?

Concrete has no reflexes, or it would be useless for sidewalks. Isn't there another way to get the idea across? 172.56.27.175 (talk) 07:49, 10 May 2014 (UTC)

Unwarranted conflict

Consider the following paragraph:

"The subject NP is shown in green, and the predicate VP in blue. This concept of sentence structure stands in stark contrast to dependency structure theories of grammar, which place the finite verb (= conjugated verb) as the root of all sentence structure and thus reject this binary NP-VP division."

In the illustration and description, there is no difference between "the subject NP" and "subject", nor between "the predicate VP" and "predicate". The older concept makes no mention of "the root" of sentence structure. So it isn't shown, at least here, that dependency structure theories of grammar "reject" anything. In fact, the illustration seems to imply that the binary division is conserved!

I can see how it might be considered unnecessary, if one puts the emphasis on finding out what the "root" is, rather than where the phrases are, but not how either theory contradicts the other. 172.56.27.214 (talk) 08:13, 10 May 2014 (UTC)

This comment is difficult for me to understand. Dependency grammars do not acknowledge the initial subject-predicate (NP-VP) division of clause structure that is the starting point for traditional phrase structure grammars. Have another look at the dependency trees in the article. Not one of them is acknowledging the initial binary NP-VP division. Not one of them shows the finite VP constituent as a complete subtree. The matter is profound, since it leads to a much different view of and approach to syntactic strucures.--Tjo3ya (talk) 17:43, 10 May 2014 (UTC)

Modern grammar example

> Who did Jim give his dog to?

Is the preposition at the end not part of the predicate? If it weren't a question, it would be "Jim did give his dog to Sally", and it being a question surely doesn't change that? Fedjmike (talk) 13:55, 13 March 2015 (UTC)

Intro too complex - need simple summary

Hi all, when coming across this article I was a bit frustrated. When you read an article on a sentence on Wikipedia it tells you it's made up of a subject and a predicate. If I'm anything like most people, I've never heard of a Predicate and want a quick snappy definition when I click on the article it links to. I notice this is the case on the Simple English wikipedia. However this article opens with "There are two competing notions of the predicate in theories of grammar.[1] The competition between these two concepts..." - completely alienating to the vast majority of the people who want to know what, in basic terms, a predicate is. Especially kids who are maybe new to the topic.

So please, please, please can someone write an introduction that isn't too technical and neatly sums up the most popular understanding of the word, so that people who are only looking for a quick summary can get on with their day! Thank you.

Whistler (talk) 23:24, 11 August 2016 (UTC)

Poor notation, inconsistent with arithmetic and computer language

I don't know whether stuff like the below is common in writings about theories of grammar, but it's not great:

Ben *reads the book*. - verb + direct object predicate

Due to operator precedence, in the expression LHS + RHS, stuff either side of the plus is evaluated and then added together. If you evaluate the RHS in that thing above, you get "direct object predicate". So you can only mean "a verb plus a direct object predicate." What you're trying to mean is "a verb-plus-direct-object predicate". So I strongly suggest that you write it that way if you want folks to understand. — Preceding unsigned comment added by 24.22.165.103 (talk) 23:07, 14 December 2016 (UTC)

Meta-informative Centering Theory page appears to be missing/'redlink', but well-documented elsewhere. Was it removed? Sobeita (talk) 13:19, 23 September 2017 (UTC)

How many types of action =

Instead of Carlson classes, consider's Schank's primitive actions. See https://deepblue.lib.umich.edu/bitstream/handle/2027.42/30278/0000679.pdf;jsessionid=B8039541E82D2A2EB49A7E14548F33E7?sequence=1 Figure 1. What is a predicate set?— Preceding unsigned comment added by 66.172.251.198 (talk) 06:07, 23 March 2019 (UTC)

Collective vs. distributive predicates - Form a line...

If a person can form a company or a government why can’t one form a line?

Like being in charge of convincing a bunch of people to queue up in a line... Mramosch (talk) 15:22, 5 July 2019 (UTC)

NPOV and sources contradicting claims

I'm restoring the article to an earlier version per WP:NPOV and WP:VERIFY. The previous text misrepresents the subject-predicate cut as a superseded view. Perhaps this was not any previous editor's intent, but it reads that way since it refers to this idea as "traditional” and the dependency grammar view as “modern”, and goes on to argue for dependency grammar in wikivoice. My proposed version addresses this problem by presenting the two views on an equal footing. The previous text also contains numerous unsourced or erroneous factual claims. For instance, the Matthews (1981) citation does not say that the subject-predicate division comes from propositional logic. I have addressed cases such as this either by removing the text or rewriting it to match the source.

In some instances, I removed erroneous text that could perhaps be restored and repaired by other editors. For instance, the previous text make the bizarre claim that the subject-predicate cut is "hardly possible” given logical representations like "laughed(Bob)". If this were in fact impossible, then no student would pass an introductory semantics course-- see pages 188-189 of the Gamut textbook or pages 16-20 of Heim and Kratzer. I suspect that the intent is something along the lines of dependency grammar allowing a tidier syntax-semantic interface (which would be awesome!) but I don't know enough about dependency grammar to rewrite the text myself. Botterweg14 (talk) 23:45, 9 April 2021 (UTC)

Hello Botterweg14, The current version does not do justice to the confusion concerning how the term predicate is understood and used in the relevant areas. Here is how the notion is defined and explained by Matthews (1997: 291) in the Oxford Concise Dictionary of Linguistics:

"predicate: 1. A part of a clause or sentence traditionally seen as representing what is said of, or predicated of, the subject. E.G. in My wife bought a coat in London, the subject my wife refers to someone of whom it is said, in the predicate, that she bought a coat in London. 2. A verb or other unit which takes a set of *arguments within a sentence. Thus, in the same example, buy is a two-place predicate whose arguments are represented by my wife and a coat.

The senses are respectively from ancient and from modern logic. For sense (1) cf. verb phrase, but that is sometimes used of a smaller unit with the predicate. For sense (2) cf. *predicator. But that is often used of the specific word in the construction (e.g. bought; moreover it is not used by logicians."

The introduction in the current version does not clearly establish that there is competition (ancient vs. modern logic) and thus confusion across the two uses of the term. The old version did do this, however. I would like to change the article back, but we should of course seek a compromise. This compromise can occur, I suggest, by including some statements about confusion due to the varying notions and by pointing to the competition between ancient and modern logic in the area.

What is the "Gamut textbook"? A direct link would be helpful. Concerning Heim & Kratzer (1998), the term predicate is not mentioned in the pages you point to (pages 16-20), and elsewhere in that book, the notion predicate is not defined or explained in a direct way, but rather the authors seem to take some vague and preconceived understanding of the concept for granted. Thus, that source is not good for helping one understand what predicates are. --Tjo3ya (talk) 10:39, 11 April 2021 (UTC)

I’d support having the lead more closely match the source you quoted. (The current text does already mention sense (2) in talking about semantic predicates and their syntactic correspondents, but perhaps the semantic side of things should get its own article to keep things tidy.) However, that source doesn’t mention any competition or confusion, and I’m not sure there is any. Dependency grammar denies the existence of predicates in sense (1), but other frameworks recognize both senses. The current version of the text mentions the disagreement about the existence of (1), including in the lead.

Here's a link to the Gamut textbook. This and the Heim & Kratzer were just for this discussion, not for sourcing a definition of “predicate” in the article. The point was just that it's well established that one can get from [S [NP Bob] [VP laughed] ] to laughed(bob), so the article shouldn't claim otherwise. Botterweg14 (talk) 23:18, 12 April 2021 (UTC)

Hello Botterweg14. Thanks for your response. I think we are close to agreement concerning the introduction. I will produce a new draft that clearly mentions both senses. I will, though, add a comment about confusion and competition. I disagree with your statement in this regard that there may not be confusion. Consider what Carnie (2013: 57-58) writes in this regard:

"In order to discuss argument structure, we first need to define some basic terms. If you took grammar in school, you probably learned that "every sentence has a subject and a predicate." Under your schoolroom definitions, the subject is usually the first noun phrase (that is, the first noun and all things that go along with it), and the predicate is everything else in the sentence. So for example, in (23) the subject is The dastardly phonologist, and the predicate would be stole the syntactician's lunch.

(23) [The dastardly phonologist][stole the syntactician's lunch].

The definition of subject isn't too bad (we'll refine it later though), but syntacticians use the term "predicate" entirely differently. The sytactician's definition of predicate is based on the mathematical notion of a "relation". The predicate defines the relation between individuals being talked about and the real world -- as well as among themselves. The entities (which can be abstract) participating in the relation are called arguments. To see how this works,..."

Carnie is clearly trying to head off confusion about how predicates are to be understood. The Wikipedia article about predicates should do the same, in my view.

Thanks for the link to the Gamut textbook. My comment is similar, though. Gamut does not directly define or explain predicate, but rather he (or she?) seems to take a vague preconceived notion of predicates for granted. Thus, the Gamut book, like the Heim & Kratzer book, is not so helpful if the goal is to establish for general Wikipedia readers (or for me, for that matter) what predicates are supposed to be.--Tjo3ya (talk) 06:38, 13 April 2021 (UTC)

Okay, we’ll talk more once you have a draft. Once again, I didn’t cite those textbooks for their definitions of “predicate”–– just to show that one particular claim in the earlier text wasn’t correct, at least as stated. Gamut is a plural “they” in case you’re curious. Botterweg14 (talk) 15:15, 14 April 2021 (UTC)

Dependency grammar

Quote: "On the other hand, dependency grammar rejects the binary subject-predicate division and places the finite verb as the root of the sentence." -- Oddly enough, the dependency tree does have a binary distinction in which the subject is opposed to a "predicate" subtree. The latter just does not contain "may"... and in this way, it mirrors the NP-AUX-VP structure of pre-IP Chomskyan phrase structure grammar. So, the text somewhat overstates the differences. Alazon (talk) 15:17, 3 October 2022 (UTC)