Semantic classifications of words

Discuss various aspects of natural language.

Moderators: Calilasseia, ADParker

Semantic classifications of words

#1  Postby seeker » Jan 24, 2014 7:18 pm

I'd like to know which semantic classifications of words (i.e., taxonomies according to their meaning) have been proposed by linguists (e.g., "words that don't refer" like logical connectives, "words that refer to things" like some nouns, etc). Also, I'd like to know if those semantic categories are shared by all languages, or not.
Can someone help me with this question, or give me some references?
seeker
THREAD STARTER
 
Posts: 868

Print view this post

Ads by Google


Re: Semantic classifications of words

#2  Postby Jef » Jan 24, 2014 7:43 pm

You can try here for a reference:

http://www.indiana.edu/~hlw/Meaning/senses.html
Jef
RS Donator
 
Posts: 1929

Print view this post

Re: Semantic classifications of words

#3  Postby seeker » Jan 25, 2014 5:07 pm

Jef wrote:You can try here for a reference:

http://www.indiana.edu/~hlw/Meaning/senses.html

The classification seems to be focused on objects (e.g. "animal, mammal, cat"). How does the author classify the words that don't refer to objects (e.g., connectives like "and", quantifificators like "all", prepositions like "before", etc.)?
seeker
THREAD STARTER
 
Posts: 868

Print view this post

Re: Semantic classifications of words

#4  Postby don't get me started » Feb 01, 2014 2:57 pm

If you want a good overview of some core ideas of linguistic and semantic categorization then I'd recommend:
'Linguistic categorization' by John R. Taylor. (OUP).
'Words in the mind' by Jean Aitchison. (Blackwell)
'Semantics: primes and universals. By Anna Wierzbicka (OUP)

My understanding is that there are a lot of fairly well defined taxonomies when it comes to the natural world, human artifacts and the like. When it comes to certain other words, they resist any taxonomy because they are linguistic 'primes', that is, they are basic concepts grounded in human cognition. Logical connectives, deictic terms and such like fall into this prime category.
don't get me started
 
Posts: 1130

Country: Japan
Japan (jp)
Print view this post

Re: Semantic classifications of words

#5  Postby seeker » Feb 11, 2014 5:41 pm

don't get me started wrote:If you want a good overview of some core ideas of linguistic and semantic categorization then I'd recommend:
'Linguistic categorization' by John R. Taylor. (OUP).
'Words in the mind' by Jean Aitchison. (Blackwell)
'Semantics: primes and universals. By Anna Wierzbicka (OUP)

My understanding is that there are a lot of fairly well defined taxonomies when it comes to the natural world, human artifacts and the like. When it comes to certain other words, they resist any taxonomy because they are linguistic 'primes', that is, they are basic concepts grounded in human cognition. Logical connectives, deictic terms and such like fall into this prime category.

Thanks! I've seen that Wierzbicka has a nativist stance regarding the set of universal primitives. Is nativism the mainstream position about those lexical meanings nowadays?
seeker
THREAD STARTER
 
Posts: 868

Print view this post

Re: Semantic classifications of words

#6  Postby Zwaarddijk » Feb 11, 2014 6:44 pm

An intriguing bit of detour is the fact that most languages historically have probably not had a word corresponding to 'and', but instead expressed conjunctions either by simple apposition or by one of the parties being marked with some kind of comitative (i.e. 'with')
Zwaarddijk
 
Posts: 4334
Male

Country: Finland
Finland (fi)
Print view this post

Re: Semantic classifications of words

#7  Postby don't get me started » Feb 13, 2014 7:16 am

Thanks! I've seen that Wierzbicka has a nativist stance regarding the set of universal primitives. Is nativism the mainstream position about those lexical meanings nowadays?


I'm not sure I'm familiar with the term Nativism when used with Semantics.I know that Wiezbicka is a central figure in Primes and Universals, but I think that there are other fields which deal with semantics from a different starting point. Probably the Cognitive Linguistic school based around the work of Langaker is another important basis for Semantic analysis.
don't get me started
 
Posts: 1130

Country: Japan
Japan (jp)
Print view this post

Ads by Google


Re: Semantic classifications of words

#8  Postby seeker » Feb 23, 2014 5:25 pm

don't get me started wrote:
Thanks! I've seen that Wierzbicka has a nativist stance regarding the set of universal primitives. Is nativism the mainstream position about those lexical meanings nowadays?


I'm not sure I'm familiar with the term Nativism when used with Semantics.I know that Wiezbicka is a central figure in Primes and Universals, but I think that there are other fields which deal with semantics from a different starting point. Probably the Cognitive Linguistic school based around the work of Langaker is another important basis for Semantic analysis.

Thanks. Do you have some references about (1) how and when children learn different classes of words, (2) which classes of words are impaired by different neurological conditions?

I've found the following classification of word classes:
"According to the distributional approach to word classes, words are grouped with certain classes mainly on the basis of their morphological and distributional behaviour: words of the same class will generally take the same sort of derivational and inflectional affixes (morphological behaviour), and will generally occupy the same positions or ‘slots’ in a sentence relative to members of other word classes (distributional behaviour)." (Evans & Green, 2006)
Then they talk about the following word classes:
* Nouns
* Verbs
* Adjectives
* Adverbs
* Prepositions
* Determiners
* Pronouns
* Auxiliary verbs
"There are several other closed-class categories that we will not discuss here, mainly including ‘linking’ categories that join sentences, like coordinating conjunctions (and, but), subordinating conjunctions (although, because), discourse connectives (however, therefore) and complementisers (for example, that in she hoped that they would be married in the snow).We will also have little to say about interjections, words like yuk! or wow! that form independent utterances and do not participate in grammatical structure." (Evans & Green, 2006)
Evans, V., & Green, M. (2006). Cognitive linguistics: An introduction. Lawrence Erlbaum Associates Publishers.
seeker
THREAD STARTER
 
Posts: 868

Print view this post

Re: Semantic classifications of words

#9  Postby don't get me started » Feb 26, 2014 12:58 pm

Sorry Seeker, I'm not really up on child language acquisition, or neurological impairments and their effects on language. I'm sure that there are multiple books and journals out there.

Regarding word classes, I remember reading one book or article years ago that argued that traditional ways of classifying words as in your list above are overly influenced by classical theories based especially on Latin, and as such are not really applicable to all languages. For example, the category of 'adverb' contains such a variety of different words as to be almost meaningless. Canonical adverbs in English such as 'slowly', 'quietly' are contrasted with such words as 'home' or 'abroad' which are much less intuitively in the class of adverbs.

In addition to this, there is the fuzziness surrounding what exactly should be counted as a word. Stand alone words like 'dog', 'go' and 'in' are all clearly words, but what about phrasal verbs in English? Is 'put on', as in 'He put his hat on' one word or two? And this is English which has a restricted morphology compared to some other languages. Japanese is an agglutinative language that builds up meanings by adding on ('gluing') suffixes.
For example the Japanese verb for 'eat' is 'Taberu'. By changing the suffix you change the meaning.
TabeTA = Ate
TabeNAI = Not Eat
TabeNAKATTA = Didn't eat
TabeRARERU = Is eaten
TabeRARETA = Was eaten
TabeTARA = If (I) ate/ If (I) had eaten
TabeSASERU = Make (me) eat

And so on. So it is theoretically possible to have a word like ' TABESASERARENAKATTARA' meaning ' If I hadn't been made to eat'
('If I hadn't been made to eat semolina at school, I'd probably like it.' There may be errors in my inflection structure but the principle stands) Does this count as one word? And if so, does it still count as a verb? And if it is a verb, can we say it is in the same category as the source verb? And if we do, are we simply hanging words from a different language on a classificatory armature from our linguistic/cultural background that may not be entirely suitable for that language?

Kind of outside the topic I know, but I do think that we have to be cautious about basing linguistic judgments on cognitive outlooks of one narrow branch of one language family. My studies of French at school and German after graduation had not prepared me for the 'We're not in Kansas anymore' experience of trying to orient myself in Japanese.
don't get me started
 
Posts: 1130

Country: Japan
Japan (jp)
Print view this post

Re: Semantic classifications of words

#10  Postby seeker » Feb 26, 2014 5:46 pm

don't get me started wrote:Regarding word classes, I remember reading one book or article years ago that argued that traditional ways of classifying words as in your list above are overly influenced by classical theories based especially on Latin, and as such are not really applicable to all languages. For example, the category of 'adverb' contains such a variety of different words as to be almost meaningless. Canonical adverbs in English such as 'slowly', 'quietly' are contrasted with such words as 'home' or 'abroad' which are much less intuitively in the class of adverbs.

OK, but I guess we need some kind of classification of words, even if it's imperfect. Do you know better proposals? For example, do you know which categories have been proposed for the classification of japanese words, from which you've shown some interesting examples?
seeker
THREAD STARTER
 
Posts: 868

Print view this post

Re: Semantic classifications of words

#11  Postby don't get me started » Feb 27, 2014 7:55 am

seeker wrote:
don't get me started wrote:Regarding word classes, I remember reading one book or article years ago that argued that traditional ways of classifying words as in your list above are overly influenced by classical theories based especially on Latin, and as such are not really applicable to all languages. For example, the category of 'adverb' contains such a variety of different words as to be almost meaningless. Canonical adverbs in English such as 'slowly', 'quietly' are contrasted with such words as 'home' or 'abroad' which are much less intuitively in the class of adverbs.

OK, but I guess we need some kind of classification of words, even if it's imperfect. Do you know better proposals? For example, do you know which categories have been proposed for the classification of japanese words, from which you've shown some interesting examples?


I don't really think that there is some alternative classification structure. I think that the way that words are classified has two aspects that need to be taken into account. Firstly, there are classifications of words that exist in one language but do not exist in another.
For example, in Japanese there are words that indicate the topic of the sentence ('WA' and 'GA'). These topic markers just do not exist in English. In Japanese one says something like;
Kenji WA Tokyo e itta. Meaning 'Kenji went to Tokyo'. (Kenji and Tokyo are proper nouns, 'e' is roughly similar to the English preposition 'to' and 'itta' is the past tense form of 'iku', 'go'.
But as you see, there is an extra word, 'WA' which is a topic marker.
So in fact, the English translation should read something like, 'As for Kenji, well, he went to Tokyo.'

The sentence 'Kenji GA Tokyo e itta' is slightly different. My understanding of the meaning of the word 'GA' in this case would change the sentence to something like, 'It was Kenji who went to Tokyo.' (I'm still very vague on this. Unsurprisingly, English speakers often have trouble with these words)

Anyways, these words are often referred to as 'particles'. I'm not sure of in some technical classification English words can be classified as particles, but as you can see, the words are not really in any intuitively accessible category to most English speakers. Undoubtedly, other languages have words whose classification eludes the intuition of Indo-European language speakers.

The second thing to take into account with word classes is that a traditional view of language was that it is primarily propositional in nature an function. Nouns describe objects in the world, adjectives describe qualities of those objects, verbs describe actions and states of those objects and so on.

But the primary function of language is actually phatic rather than propositional, that is, we humans use language primarily as a means of creating and maintaining social bonds rather than making truth statements about the external world. So some words have a meaning and usage which is basically grounded in the discourse.

For example, the adverb 'Well' is used to describe an action in positive terms. 'She sang well' or 'Well done!' However, the most common usage of this word is in a discourse marking function. It is polysemous but has one function of turn initial placement in a second turn which follows a prior turn which was a question, and the function is to indicate non-straightforwardness in the reply.
E.g.
A: When does the party start?
B: Well, we're telling people to arrive between 9 and 10, but we'll probably start eating once most people are there, so you can just show up anytime from about 8:30.

In this case the interactive function of the word 'Well' is paramount (It is showing the questioner that although the question was transactional, the answer is more interactional) and to describe it as an adverb and leave it at that really doesn't indicate what is doing in this sentence.

So, to sum up: There are classes of words that may exist in one language but not in another and there are words whose main function is interactional and therefore only poorly described by traditional word classes.
don't get me started
 
Posts: 1130

Country: Japan
Japan (jp)
Print view this post


Return to Linguistics

Who is online

Users viewing this topic: No registered users and 1 guest