Talk:Branching (linguistics)

Restored original content
I've just reverted to the original version of the page. The article needs an intro for context, and especially to specify that the noun phrase as defined is an English noun phrase. As for English branching: overwhelming use of prepositions, SVO order and auxiliary-verb order mean it's right-branching. Left-branching only occurs at the noun phrase level and less strictly in the adverb-verb phrase order. --Pablo D. Flores 01:01, 10 Jan 2005 (UTC)

Head-last/first
Before finding this page via Japanese grammar I had never heard of the "branching" nomenclature. Perhaps some mention that "left-branching" and "right-branching" are equivalents to "head-last" and "head-first" would be beneficial.


 * I think I've fixed it. I've also moved some content to the first paragraph in order to establish the context sooner, otherwise the article takes too much time to get to the point... --Pablo D. Flores (Talk) 22:09, 11 October 2005 (UTC)

Left branching in Spanish.
Since there seems to be a disagreement here about the occurrence of left branching in Spanish, and edit summaries aren't a great place for explaining positions, let's discuss here.

In Spanish, object personal-pronouns always precede their verbs (lo hago), unless their verbs are infinitives, gerundios, or imperatives (hacerlo, haciéndolo, hazlo). (Objects also precede verbs in certain other contexts, such as in "¿Qué hace?" and "No sé qué hizo", but these aren't really examples of left branching.) Additionally, sometimes an object can be moved forward in order to emphasize it, but I think this is actually an example of left dislocation, and at any rate, saying "Spanish [...] allows direct objects before verbs for emphasis" is misleading because this is not the most common reason for an object to appear before its verb.

Also, "Spanish, which is overwhelmingly right-branching, allows [...] adjectives before nouns for stylistic reasons, although with certain nouns this change in position might suggest different connotations" just seems completely wrong. There's a meaning change, not a stylistic one, in moving an adjective before a noun: it indicates that the adjective is non-restrictive: "ADJETIVO SUSTANTIVO" means "SUSTANTIVO, que es ADJETIVO", while "SUSTANTIVO ADJETIVO" means "SUSTANTIVO que es ADJETIVO". Additionally, with certain adjectives (not nouns!) there's a separate change in meaning when this happens; for example, "SUSTANTIVO grande" means "big NOUN", while "gran SUSTANTIVO" means "wonderful NOUN" (so that both come out meaning "great NOUN", heh), but this is not the most common reason for an adjective to appear before its noun. (It is a genuine example of left branching, so I'm not opposed to mentioning it, but the current phrasing is misleading both in its description of it, and in making it sound like it's the most common reason for an adjective to precede its noun.)

(Disclaimer: I'm not a native, or even particularly good, Spanish speaker, but I'm fairly certain I know what I'm talking about here.)

Ruakh 05:20, 10 January 2006 (UTC)

More languages and classification
I'd like to see the branching tendancies of more languages and language families be talked about, for example, the insular celtic, semitic, and sino-tibetan languages. The only problem is classification, since a lot of languages seem to be shifting from one to the other (mostly left to right). Is there any language which is as rigidly right-branching as japanese is left? —The preceding unsigned comment was added by Bob A (talk • contribs).


 * Hmm. Hebrew is generally SVO (or VO — though this is fairly flexible anyway), it uses prepositions almost exclusively (including a number of prepositional prefixes, while I can only think of one postpositional suffix), and it puts nouns before adjectives (and often before determiners, even), so I guess it's fairly consistently right-branching. (comment continues below)


 * True, but it still has some postfix inflection, which i believe is left branching. Bob A


 * Are you saying that given an inflected word, the inflection is the head and the rest is the dependent? Ruakh 17:50, 9 May 2006 (UTC)


 * Yes; it seems that way to me. For example, in french, plural nouns are indicated by the prepositions les and des. and tense/aspect is indicated by the words for have and go. I'm not sure, though. Bob A 19:29, 9 May 2006 (UTC)


 * That said, I'm not sure how much it would contribute to the article to discuss the branching tendencies of so many different languages; I think it would be more interesting to discuss wider trends, such as how well SVO/VSO-ness correlates with prepositionality, how well each of them correlates with noun-adjective word order, and so on. (I'm sure such research must exist, as it seems necessary to even justify the existence of the terms right- and left-branching.)


 * Ruakh 21:59, 8 May 2006 (UTC)

Adjective placement in English
One piece of information in this article is obviously incorrect. It states that an adjective can only be added to a bare noun in front of the noun. What about "galore" as in "I saw bargains galore at the mall"? —The preceding unsigned comment was added by 207.232.181.155 (talk • contribs).


 * Good call. I've fixed that now. —RuakhTALK 18:05, 21 December 2006 (UTC)

X-bar theory
It seems logical to add an explanation branching in terms of X-bar theory and to introduce specifiers, adjuncts, and complements. -- Beland 19:25, 21 April 2007 (UTC)

Transformations
There is no mention in the article of transformational grammar. I'm wondering if in some cases the branching tendency is consistent over the deep structure, but appears inconsistent over the surface structure. Or do transformations not affect branching order? I would be more comfortable with the conclusions of the article if references were supplied. -- Beland 19:38, 21 April 2007 (UTC)

Relationship with word order
It would be very illuminating to explain, preferably showing some example trees, how branching order gives rise (or is different from) to various word orders, e.g. SVO. -- Beland 19:46, 21 April 2007 (UTC)

Articles
According to the entry an article is a kind of dependent and is therefore subject to branching rules in the same way that adjectives are. However, The Universals Archive says this:


 * There is a statistically significant correlation between the order of article and noun and the order of object and verb. The direction of this correlation, however, is the opposite of what traditional assumptions might lead us to expect. If one adopts the traditional assumption that articles are modifiers of nouns and the common view that modifiers tend to precede the noun in OV languages and to follow in VO languages, then we would expect articles to precede the noun more often in OV languages than they do in VO languages. But the correlation is in the opposite direction.

This suggests that the page as it currently reads is wrong. I suggest that the page be modified to reflect the quote above or that discussion of articles be removed altogether. I imagine our decision will depend on what other references we can find, which is why I'm asking for feedback before making any chances. -- SgtSchumann (talk) 21:33, 15 March 2009 (UTC)


 * The current version of the article says, Though articles [] have been thought of as the dependents of nouns, it is now thought that articles and nouns form a dependent phrase, in which the article is the head., which seems to suggest that in the house, the is the head and house is only a modifier of the. This is counterintuitive and not supported by the source given. The source (citing some "Abney (1987)" -- maybe "The Noun Phrase in its Sentential Aspect") talks about seeing the as the head of its own subphrase, D, that could contain other modifiers of the noun (including other nouns), but it doesn't say that the is the head of the whole DP the house. The unexpected direction of a correlation should be noted as an open question, but not lead to assertions that are not supported by research. WP:OR --88.73.26.25 (talk) 21:31, 13 August 2011 (UTC)

Japanese "rigid model" for strictly left-branching languages?
Granted, Japanese is a good example of this kind of language, though it has at least one exception to the left-branching rule: numerals follow nouns in Japanese. Example: hon nisatsu="two books" lit. "book two-counter word". —Preceding unsigned comment added by 70.122.246.31 (talk) 23:28, 10 April 2009 (UTC)


 * Actually the article says, "Practically a model for rigidly right-branching languages," which sounds about right to me. -- SgtSchumann (talk) 02:06, 11 April 2009 (UTC)

No, it says word for word "practically a model for rigidly left-branching languages", though it is not necessarily. Japanese is largely left-branching(essentially the modifier precedes the modified, objects appear before verbs, etc.), though its nouns precede the numeral, which is a characteristic of right-branching languages. —Preceding unsigned comment added by 65.65.52.43 (talk) 22:44, 11 April 2009 (UTC)

Anglocentrism
Am I the only one who thinks that the terminology "head-first" and "head-last" is terribly Anglocentric? It is true that, since they are used, they should be covered here, but are there more linguistic, dare I say realistic terms available which Wikipedia could more strongly endorse? I could be knee-jerking, but there's no explanation given for why we should call the ordering of right- and left-branching languages by those names. I also think that the demonstration of left-branching is rather weak given the fact that the Hilary Clinton sentence can be ordered in numerous equally viable ways in English; not knowing Basque, I can't say whether the same holds true. I'm sure, in any case, that we can do better with what's being presented on left-branching. [I actually just realised that my tendencies in ordering English (my native tongue) and reading differently ordered English are probably affected by my Turkish studies. Dunno if that's relevant to the opinions I just expressed.] this raven is icy (talk) 09:59, 25 June 2009 (UTC)


 * This is interesting. It seems to me that the terms right-branching and left-branching are ethnocentric. After all it is not the case that all languages are written in such a way that they would be head-first if they were "right-branching" or head-last if they were "left-branching". I suppose that head-first and head-final could also be ethnocentric if the first and the last were taken to indicate spatial relations, but I've always taken them to indicate temporal relations. Is there something more to this that I'm missing? -- SgtSchumann (talk) 19:02, 8 July 2009 (UTC)


 * (i think), it is not ethnocentric. do you know why head is named head? because it is main part of a 2-part phrase. (what problem do you see)? --Qdinar (talk) 15:10, 23 July 2016 (UTC)

More examples
This is an interesting article but to improve the understanding of head first and so on there should be more examples given (in English) in the first couple of paragraphs. I am trying to think of some! —Preceding unsigned comment added by Bewporteous (talk • contribs) 19:18, 21 February 2010 (UTC)

This sentence from the article does not make sense
Languages like English and German - though regarded as being right-branching because their main verbs precede direct objects - place prepositions, adjectives of location, and numerals before their nouns.

This business of "adjectives of location ... before their nouns" is true, but it excludes one whale of a lot of other adjectives that are also placed before the noun that they modify, but which are not "adjective of location". For example: a white horse, a sharp ax, a heavy load, an orange tree, a fast rocket, ancient Egyptian, Ancient Greek, foggy morning, independent country. In other words, just about any adjective (a word, not a clause, or a phrase) in English comes before the noun or pronoun that it modifies, and not just adjectives of location (such as "Egyptian mummy", "British tea", "Japanese car", "American people", and so forth.)98.67.170.121 (talk) 11:49, 6 July 2010 (UTC)

Binary vs. n-ary branching
Someone seems to have an academic bee in their bonnet due to the overstructuring (?) of their systax trees. Is there any sort of consensus in linguistics that X-Branching trees are "confusing"? --- 173.74.176.81 (talk) 01:21, 21 August 2012 (UTC)


 * I'm not sure what the question is here. I found X-bar trees confusing when I first learned about them many years ago, and today I still do not see that the "tall" trees resulting from the X-bar schema are an insightful way to render syntactic structure. I think many syntacticians would agree, since X-bar structures have long since been abandoned by many working within the Chomskyan tradition. --Tjo3ya (talk) 04:15, 21 August 2012 (UTC)

Determiners as heads
Many modern theories consider determiners, rather than nouns, to be the heads of phrases such as 'the house'. In light of that, I revised the example sentences, but I neglected to notice the tree diagrams using the older structure. I hope this inconsistency is fixed at some point; sorry. Feel free to revert my changes. TheNyleve (talk) 21:21, 4 February 2014 (UTC)


 * Yes, I reverted your changes. The DP-hypothesis is unique to GB/MP. It does not extend beyond these theories. See the article on determiner phrases for discussion: https://en.wikipedia.org/wiki/Determiner_phrase . --Tjo3ya (talk) 00:58, 5 February 2014 (UTC)


 * True, but to be theory-neutral I would recommend not including English examples with NP/DP. I did not, for example, add DP to the other example list. It would also be OK to add a note or something that this analysis isn't shared by all linguists. TheNyleve (talk) 04:57, 5 February 2014 (UTC)


 * do you mean that there are theories that say that "the" is head (main part) and "house" is non-main/helper/modifier part in "the house"? --Qdinar (talk) 15:04, 23 July 2016 (UTC)

all indo-european are like english?
are all indo-european languages like english, ie with adjectives & adverbs & articles earlier (at left in text) than head, and objects and prepositional phrases ("of me") are later (at right side in text) than head (where head is verb or preposition or main part relatively to prepositional phrase, for example "of you" relatively to "book" in "book of you"), and сonjunction phrases (i mean, for example, "and i go") are later than main phrase ("you come and i go") and additional phrases/sentences, joined with conjunction, ("i go") are later conjunction ("and")? and family names are later than first names? in english and russian this is so. --Qdinar (talk) 15:23, 23 July 2016 (UTC)

i see there are some exceptions in english and german: http://linguistics.stackexchange.com/q/1762 --Qdinar (talk) 15:55, 23 July 2016 (UTC)

are all uralic and altaic and dravidian languages like turkish?
as i see, in turkic languages head is always later than dependent part, except conjunctions and family names. (because conjunctions are from persian and arabic). are all uralic and altaic and dravidian languages like this? --Qdinar (talk) 15:28, 23 July 2016 (UTC)

i see there are exceptions in balto-finnic languages of uralic languages: they have some prepositions: http://linguistics.stackexchange.com/a/1774 --Qdinar (talk) 15:54, 23 July 2016 (UTC)