Syntactic and lexical categories

From Helpful
(Redirected from Auxiliary verbs)
Jump to navigation Jump to search

Language units large and small

Marked forms of words - Inflection, Derivation, Declension, Conjugation · Diminutive, Augmentative

Groups and categories and properties of words - Syntactic and lexical categories · Grammatical cases · Correlatives · Expletives · Adjuncts

Words and meaning - Morphology · Lexicology · Semiotics · Onomasiology · Figures of speech, expressions, phraseology, etc. · Word similarity · Ambiguity · Modality ·

Segment function, interaction, reference - Clitics · Apposition· Parataxis, Hypotaxis· Attributive· Binding · Coordinations · Word and concept reference

Sentence structure and style - Agreement · Ellipsis· Hedging

Phonology - Articulation · Formants· Prosody · Sound change · Intonation, stress, focus · Diphones · Intervocalic · Glottal stop · Vowel_diagrams · Elision · Ablaut_and_umlaut · Phonics

Speech processing · Praat notes · Praat plugins and toolkit notes · Praat scripting notes

Analyses, models, software - Minimal pairs · Concordances · Linguistics software · Some_relatively_basic_text_processing · Word embeddings · Semantic similarity

Unsorted - Contextualism · · Text summarization · Accent, Dialect, Language · Pidgin, Creole · Natural language typology · Writing_systems · Typography, orthography · Digraphs, ligatures, dipthongs · More linguistic terms and descriptions · Phonetic scripts

This article/section is a stub — some half-sorted notes, not necessarily checked, not necessarily correct. Feel free to ignore, or tell me about it.

Varying names and concept

The core concept here is indicating groups of words that have similar grammatical properties.

For example, verbs all function more or less the same, and when they don't we qualify how they vary.

The name for this idea is variably.

  • Part of Speech (POS)
  • lexical category
  • lexical class
  • word class
  • grammatical category

...and more.

Sometimes even with topic-specific and even conflicting definitions.

According to some definitions, lexical category only deals with nouns, verbs, adjective and, depending on who you ask, prepositions.

Under such definitions, look to to word class for all the smaller ones.
this seems to be part of an attempt to a relatively modern attempt to replacing the earlier and vaguer part of speech, though this doesn't seem to have taken off in a large way, possibly because POS is still seen as an umbrella term?

And then some argue that you can have:

Lexical category is category we ascribe to a lemma in use(verify), grouping of words by their syntactic and/or morphological behaviour.

Phrasal category is the same idea, but indicating the function of a phrase. Examples include noun phrases and verb phrases.

Syntactic category can include both lexical categories and phrasal categories.

And then there are further grammatical constructions. Say, what is an apposition? Is thjat wider

When you're diving into the theory, these terms matter to know - but it is also useful to know that the typology of these terms and similar has always been a damn mess.

If you're using software that labels words with these groups, it basically just doesn't matter. And is usually called POS.

Varying typologies

The more basic list includes:

Some catch-alls

Some specifics:

It should be pointed out that

  • any such class/category is much more of a description of use in a sentence, than of a word's intrinsic properties
in that one word can be used in multiple classes
and in that not all words in a category work exactly the same, particularly in the broad categoies

  • the most common typology is tuned for Indo-European languages
largely because that's where such linguistics were focused first

  • such class/category (and dependency relations like 'we verb that noun') alone is a far-from-complete model of sentence structure
in that there are more things that need to work out, many of them could be considered related to case, modification, and agreement]


  • Many languages have a bulk of words that clearly fit into only one or two major categories (that may be somewhat specific to that language), such as nouns and verbs, but there are often different subdivisions based on language, preferred detail, and the model as a whole may differ.

The variation in language-specific tagsets illustrates some of this.

  • Analytic languages do not always mark words as belonging to a category, instead leaving interpretation (and therefore also the lemma) mostly or entirely up to grammar or other usage context. This also leaves some more room for more creative usage.

(Even synthetic languages often have some creative room. For example, various nouns may be verbs may be usable as verbs and nouns (resp.) to some degree, as in e.g. "They didn't language that proclamation very well," or in "The hows and whys." (how and why are purely nouns))

It may be interesting to see the counts of words within the categories of common words/lemmas in the vocabulary of a language. In English you will probably see that


A verb is the lexical category that in most languages expresses action, and often occurence, existance and the likes.

Some people like to make further distinctions like: Main verbs refer to the verb describing the primary activity.

Action verbs refer to transitive verbs, and are named such because they tend to act on, do something to an object.

Modal verbs refer to verbs that indicate ability, obligation and such - can, should, must, may,

Linking verbs refer to copulae.

Verb classes

A verb class is a group of verbs that share a specific set of linguistic properties, such as relatization of arguments and their interpretation.

In inflected languages they usually share a very similar conjugation pattern.

Examples often use semantically related verbs, such as hit and break (see Fillmore (1970) The Grammar of Hitting and Breaking), in part to demonstrate how words you think of as similar diverge in behaviour when you play with transitivity, grammatical cases, and such.

English has over 250 verb classes (verify), Latin has four, Japanese has two.

In many cases, languages with fewer verb classes are much more regular (in general), and also tend to be easier to learn.

Regularity (property)

Whether a verb is regular or irregular describes whether it does or doesn't follow the language's general rules for conjugation.

Regular verbs follow them, irregular verbs do not and are exception cases one has to remember to use correctly.

The most common verbs in a language are often irregular, while less commonly used verbs are often very regular.

(Ir)regularity is a somewhat fuzzy and subjective quality, and depends on the language itself. Clearer cases are often agreed on, while the number of border cases may well be argued about.

See also the related concepts of weak and strong inflection.

Transitivity (property)

When you have a subject and...:

  • object, it is intransitive, for example 'I am,' 'It rains', 'He sleeps,' 'They run'.
  • object, it is transitive, for example 'I am me,' 'He cooks food,' 'They hit me'.
  • ...two objects (one direct, one indirect), it is ditransitive, for example 'I gave Brak a banana.'

A less common way of describing that same transitivity is valency, which counts the number of arguments to a (verbal) predicate, including the subject. An intransitive verb would be monovalent, a transitive one divalent, a ditransitive one trivalent.

Intransitives with dummy subjects are sometimes called avalent, such as with the subject in 'It rains' - there is no specific thing, person or other actor that is raining.

Verbs such as this, that never take a real subject, are sometimes called impersonal verbs.

Transitivity is somewhat relevant to word order typology, since different rules apply when you don't have exactly one object.

When talking about direct and indirect objects, you also involve cases.

Languages like English have a good number of verbs with flexible transitivity. When a verb can act as more than one of these it is technically called ambitransitive.

Tense (property)

The amount of tenses varies per language. Some languages have a few or even none, Germanic languages often have roughly the below set, some languages more than this.

  • Simple past, eg. "I went.", "I took" (also known as preterite, past indicative, past historic)
  • Simple present, eg. "I go.", "I take"
  • Simple future, eg. "I will go.", "I will take"
  • Present continuous, eg. "I am going.", "I am taking"
  • Past perfect, eg. "I had gone."
  • Present perfect, eg. "I have gone."
  • Future perfect, eg. "I will have gone."

Many languages tend to not inflect for all of its tenses, and use auxiliary verbs to mark certain forms.

Also, you may see a lot of overlap. For example, in English the simple past and part participle are frequently take the same form.

Infinitives, participles, and other derivatives

This article/section is a stub — some half-sorted notes, not necessarily checked, not necessarily correct. Feel free to ignore, or tell me about it.

  • Finite verbs are inflected, and often usable as main verbs in (independent) clauses, regularly in the form of an assertation.
  • Non-finite verbs are not inflected, and include verb derivatives, and often means the word can be used both as an active verb and in another
    • Infinitive: root of a verb - the least marked form, the one you would use in 'to [...],' e.g. 'to sleep',
      • present infinitive, eg. "I like to go"
      • perfect infinitive, eg. "I like to have gone"
      • Not all languages have the distinction of full and bare (with and without the 'to')
      • An infinitive phrase is one that starts with to in this sense.
    • Participle: adjective derived from a verb, eg. 'The breaking window', 'The done deal'.
      • Present participle, as in eg. "the confusing professor"
      • Past participle, as in eg. "the confused student"
    • Verbal noun variations, such as gerunds. See noun.

A verbal refers to non-finite verbs. The name comes from traditional classification into verbal nouns, verbal adjectives, and verbal adverbs.


See also:

Compound verb

Auxiliary verbs

Auxiliary verbs (sometimes 'helping verbs') are verbs that assist a main verb (also 'lexical verb', sometimes 'full verb').

For example, working with the finite main verb 'write', you can vary that into 'have written,' 'have been written,' 'am writing' and so on.

English auxiliary verbs

There are five types of auxiliary verbs:

  • Passive auxiliary verb
  • Progressive auxiliary verb
  • Perfective auxiliary verb
  • Modal auxiliary verbs cannot be finite verbs. For example, can/could. There are nine of these in english.
  • Dummy auxiliary verb

Various auxiliary verbs, e.g. 'be' and 'have', can be used both as regular and as auxiliary verbs. Template:Verb


A noun (or noun phrase) refers to a person, place, thing, event, substance or quality.

Noun properties (/classifications)

Properties are cognitively used to disambiguate on a lexical and semantic.

For example, specific transitive verbs may or may not take or produce animate objects.


A noun can be posessive or not, in English referring to an apostrophe-s tacked on the end, like in bird's and and birds'.

How linguistic analysis treats apostrophe-s varies a little. In some ways this is a simple inflectional suffix, in some ways the result acts like a pronoun form (but is not equivalent), and apostrophe-s is considered by some taggers as a part of speech upon itself, which is generally not so practical.

On plurality and countability

In languages that mark for plurals, nouns are interpreted (and almost always marked) as either singular or plural,

For example English declines its plurals to show their number (one hat, two hats), which mostly happens with a predictable and simple affix pattern.

Other languages show or imply number in other ways.

Not every English noun has a plural form.

This relates to a word's countability:

  • A Countable noun, also count noun, is one that has single and plural form, e.g. chair, chairs
answerable to a 'how many?' question.
  • Non-countable noun, a.k.a. mass noun: those that does not pluralize
answerable to a 'how much?' question.
often practically because you can't discern units of what you are referring to, eg. furniture, water, wine.
these words may still be used in plural form, but then not to refer to much of it, but to refer to a non-concrete type, as in e.g. 'the wines of france'
  • Collective nouns refer to groups, such as flock, herd, bunch, audience and class.
Collective nouns are usually countable themselves so appear in both singular and plural form, but aren't always countable

A noun can sometimes be classified as more than one of each, though most are just one of countable, mass or collective.

In some cases, different of these forms are homographic: 'stone', for example, can be seen as countable objects as well as a mass noun indicating the material.

There are also the plural-only cases, such as pants, glasses, scissors. Which in some cases may be homographic with something else.

There are cases with some argument. For example, software is usually treated as a mass noun, but some people use it as a countable (collective?) one.

Related grammatical rules

Countable collective nouns nouns can often be used in more than one form. For example, in "The committee has decided", you use 'the committee' in singular form to refer to the group as a concept.

"The committee have decided" is also correct, and 'the committee' here suggests plurality and thereby the interpretation "The members of the committee have decided." However, it seems a generally avoided form, probably because it is harder to parse; you have to go figure out just how collective and countable the noun/NP is.

It is also easy to tangle yourself up in agreement this way - "The committee have taken their seats," is arguably nice and unambiguous form, but try using 'has' and you're in trouble.

Proper versus common

Proper nouns are the names of of specific concepts, exemplars, or perhaps more usually specifically indicated physical things, places, or people.

For example, Paris will usually be understood as the Paris in France. (Or the one in Texas. Or a Shakespeare character, but is a proper noun in all of these cases.)

Proper nouns are opposed to common nouns, which can refer to many objects.

For example, a door could be any door in the world. (Often seen with limiting modifiers like a/an, some, every.(verify))

Some nouns can act as both a proper and a common noun, but most are used as just one.

Proper nouns are not generally descriptive; they are expected to be known.

Personal names (a subset of proper nouns) are almost per definition non-descriptive words, even if the words themselves could be interpreted as such.

Most latin-written languages currently capitalize many (but far from all) of their proper nouns (examples: Big Ben, Jakarta, black death), but basically never common nouns (examples: raindrop, rose, whisker, kitten).

There are a few exception languages, for example German, which capitalizes all its nouns.

Proper nouns do not describe properties, common nouns may, although this is where things start to get into hairy philosophizing.


Names can be grouped under proper nouns, and are regularly considered a special type of them.

Note that names do not quite act like regular words/nouns in various ways.

For example, they are treated as non-semantic words. In translation, they are often only phonetically adapted. They are often transliterated and converted to use another language's diphone set, then written using the target language's orthography.

Certain inaccuracies may be avoided by translators, which is one of the causes of there being many different written forms of a translated name.

More notes on personal names

Given name or first name (in some contexts christian name) distinguishes people with the same family name. Given to a person instead of inherited.

Family name (also surname, last name), inherited from family, often shared by family except by taking a name (often in marriage, or related to parents). Usually constant for many generations.

Legal name refers to the full name appearing on official documents. One's legal name can change

Name change typically refers to changes to one's legal name. Changing one's surname is often most accepted at marriage (otherwise legally bothersome, sometimes even impossible). Changing one's given name is more easily accepted, though also often less necessary as people will easily call you by a name that isn't your legal given name.

Married name refers to the name after change at marriage - often by choice.

Maiden name refers to one's name before marriage (or, if you don't take the other's name, after too). The term is applied when women take another name, but the concept can apply in any direction.

See also:

Concrete versus abstract

You can distinguish between concrete and abstract. (This is mostly applicable to common nouns, since proper nouns have more of a tendency to be concrete)

  • Concrete nouns refer to sensable (tactile, etc.) things.
  • Abstract nouns refer to ideas or qualities, e.g. height, hearing, February, and truth.

Animate versus inanimate

Matters as one of many properties that work in disambiguation.

Worth mentioning as various of the most common verbs may select for animate nouns.

Related things

A proper noun is a name for a specific place, person, or thing.

Many languages capitalize proper nouns opposed to a common noun, which name classes of things.

A pronoun can replace a noun (often with some other change, e.g. in terms of a determiner).

say, in a longer language we use we, you, they, he, she, it, (and some variant forms)

to refer to things or people that were previously named specifically, largely to avoid naming that thing again and again.

A Gerund is a verbal noun that is functionally a noun, but locally looks like and interacts as verb, for example through being adjustable by an adverb, or having a subject.

Example: 'editing' in "[Editing this article] is easy." (the first three words are a hen considering the first three words an NP to 'is.')
Many gerunds are formed by adding '-ing,' but this is not a conclusive test.

A non-finite verb is another type of verbal noun; it does not necessarily correspond with the subject in various ways.

A noun adjunct, a.k.a. attributive noun or noun premodifier, is a noun that acts as an optional modifier on another noun. Optional in that removing it does not change the sentence grammar. For example, 'chicken' in a sentence mentioning chicken soup does not change the structure, and only has an effect on .

An adjectival noun refers (fuzzily) to any mixed adjective-noun function. It usually refers to an adjective used as a noun (such as the poor), but also occasionally to noun adjuncts.

verbal nouns - nouns that come from a verb or from a verb phrase, that retain some verb-like properties

deverbal nouns like a verbal noun, but basically autonomous.


  • example: `being' in The question of being is a verbal noun (verify)
  • Fencing is fun - verbal noun; verb-like in that you can replace it by the infinitive "to fence")
  • The white fencing is nice - deverbal; you could replace this with a noun, like "bench")

There are semantic variations you can ascribe within deverbal nouns, including:

  • Agent nouns - invader, singer - the agent of the action.
  • Patient nouns - draftee - the object of the action.
  • Result nouns - dent, bruise
  • Ability nouns - speech, as in "She regained her speech."
  • Manner nouns - walk, as in "She has a funny walk."

Cases and declension

Declension refers to the general concept of a noun changing into a form (declining into a case) when used in a particular role, such as posession or any of a few dozen other reasons.

For example, a noun in dative case most often indicates that a noun, NP or PP describes someone to whom something is given. For example, in I give you a book, 'you' is the dative (besides being one of the ).

In I give a book to you, the PP 'to you' is dative.

In English there is no particularly marked dative case other than in a few expressions, so linguists do not necessarily use this term.


See Pro-forms


Adjectives modify the semantics of nouns or noun phrases, usually making it more specific, such as making a 'ball' a 'blue ball'. (Compare with adverbs, which are used to modify most everything not a noun).

In English, adjectives do not designate case, number or gender.

  • Comparative adjectives introduce a (ranked) comparison of two situations.

The following are similar to the pronouns used in the same way:

  • Demonstrative adjectives are functionally identical to demonstrative pronouns, but directly modify nouns or noun phrasees.
  • Interrogative adjectives
  • Posessive adjectives
  • Indefinite adjectives

See also:


This article/section is a stub — some half-sorted notes, not necessarily checked, not necessarily correct. Feel free to ignore, or tell me about it.

In linguistics, "nominal" points out that in many languages, nouns and adjectives tend to share a lot of syntactical, and possible morphological properties, which also means they share patterns of agreement, that you can study together.

This may further extend into pronouns and possibly numerals.
in European language this seems to originate mostly in the often Latin-based grammar, and its productive affix-based marking.


Roughly speaking, adverbs modify (often describes or augment augment) anything not a noun. (Contrasted with adjectives, which modify nouns.)

Adverbs tend to modify verbs, adjectives, other adverbs.

Some can modify phrases, clauses, and sentences.

Adverbs often add some detail of time, place, cause, manner, degree, etc.

Some words may be used both as adjectives and adverbs, particularly in languages that do not mark one or the other.

In English

Many adverbs are marked with the -ly suffix.


  • He quickly came over and explained at length how amazingly dull the movie was.
  • The meeting went well, and the directors were extremely happy with the outcome.
  • I often have eggs for breakfast.
  • Crabs are known for walking sideways.

The surface form of some of these non-noun modifiers (particularly the -ly form) is different from that of noun modifiers (adjectives). This helps avoid some ambiguity by reducing the amount of readings of sentences.

See also


This article/section is a stub — some half-sorted notes, not necessarily checked, not necessarily correct. Feel free to ignore, or tell me about it.

Determiners modify nouns in specific ways.

They are similar to adjectives in that respect, and the two are sometimes grouped into the same general category for convenience, even if they are grammatically distinct in a language (such as in English).

'Determiner' itself a term grouping a number of categories, and includes:

  • articles, such as the definite article the and the indefinite articles a and an.
  • possessive determiners, such as my, our, your, her, his, its, their, whose


Pronouns do not follow any strict typology, mainly a number of properties that can be seen as various groups (that occasionally interact).

Pronouns include (these cover most common cases, but not all):

  • Indefinite pronouns are references that generalize or are not direct or decided. This includes anyone, none, one, more, each, neither. Plural indefinites would include few, both, some, several, and many.
  • Demonstrative pronouns point to some identifiable concept or things: this, that, these, those, such, as in "Is anyone using that fish?" or "Such is my belief." (cf. determiners)
  • Personal pronouns cover references to the grammatical persons. Their form shows what they are used for, subjective (I, you, she, he, it, we, they) or objective (me, you, her, him, it, us, them). Note, however, that 'it' is often non-referential.
  • Posessive pronouns are forms of personal pronouns that mark ownership, wuch as mine, yours, hers, his, its, ours, theirs. (cf. determiners)
  • Interrogative pronouns are wh-type words that are used to ask a question, whether it is in a main clause (e.g. "Which one is it?") or a dependent one (e.g. when it is used as introduction, as in "I wonder which one it is?")

Pronoun properties include things like:

  • Relative pronouns are wh-type words that connect (and start) relative clauses to others, such as that in "This is the fish that no one ate."

They can also be subcategorized by case and by whether it is personal:

  • Objective case: whom, whomever, that (personal), which, that (non-personal), and also where, when, why.
  • Subjective case points to the things they refer to: who, whoever, that (personal), which, that (non-personal), e.g. "Is anyone using that fish?" Which refers to things, who/whom/whose refers to people, and that may refer to either.
  • Posessive case, whose

  • Reflexive pronouns indicate that the subject and object are the same: "I asked myself [...]". Includes myself, yourself, himself, herself, itself, oneself, myselves, yourselves, themselves. They appear after the noun it refers to.
    • Intensive pronouns refers to apposative reflexives, used to emphasize the antecedent, e.g. "I myself did this."

People misuse reflexive pronouns in a few ways

  • My friend and myself vacuumed the place.
    • the argument is the same as the me/I case: This is two sentences in one, and 'myself vacuumed the place' isn't right - it should be I.

Some people object to "young people like myself", but I can't figure out why.

The I, me, myself problem


This article/section is a stub — some half-sorted notes, not necessarily checked, not necessarily correct. Feel free to ignore, or tell me about it.

Definiteness is about how specific a reference is.

Many languages mark definiteness in a sentence, via specific determiners or more general articles. For example, talking about a book could be any, talking about the book is probably a specific one.

There may be language-specific details to this.

For example, in Dutch, 'de' (the) is typically a definite article - but sometimes not. For example, in "Ik heb bij de Ethiopier gegeten", 'de Ethiopier' is not a reference to a particular ethiopian as would be the literal interpretation, but to a habit of referring to restaurants this way.


Pronouns can often be used instead of the subject of a verb, object of a verb, and as object of a preposition.

As a referrer, pronoun-referent resolution depends largely on agreement, often in gender and number.

There are various properties and distinctions of pronouns, sometimes introduced as classifications. Some distinctions of pronouns are relatively independent of others, some may apply to various pronoun types, and some pronouns can be used in more than one sense.

In more official writing, genderless pronouns have seen argument. Historically, 'he' has been used where genderless pronouns would be more correct, politically as well as generally. Solutions like he/she, s/he, one, it, and making up words have not been generally satisfying (in some cases because it the word is always neutral and therefore unconditionally ambiguous). It seems that the best current solution is the they (used singularly and plurally), even if that mildly messes with number agreement.

See also


This article/section is a stub — some half-sorted notes, not necessarily checked, not necessarily correct. Feel free to ignore, or tell me about it.

Definitions may focus on one or more of

that it's a function word that helps add meaningful context
that it does not fit one of the existing lexical categories
possibly a word that might normally fit one, but does not in this particular use
that it cannot be inflected so has a fairly fixed form

As such, it's a bit of a catch-all term that is vague definition that varies in which potentially-distinct thigs it might include.

Say, some include prepositions, even articles, others treat it as a 'miscellaneous' category.

People may identify specific variants, e.g. phrasal praticles

Phrasal verb are often fairly-idiomatic combinations of a verb plus particle. Fill in, traveled to, turn up, put out, break down, took off, went up (and here mostly adpositions).

See also:


This article/section is a stub — some half-sorted notes, not necessarily checked, not necessarily correct. Feel free to ignore, or tell me about it.

An article is a determiner that combines with a noun phrase.

In many languages it assigns a definiteness to the noun and the reference it makes, and in some languages it also marks things like gender, number, and case - and the article's (correct) form may depend on the properties of the noun it's on.

A number of languages has no articles.

An indefinite article is used to refer to something as part of a group. In english this would be a/an, or negative forms.

Example: A robot is systematic.

A definite article makes a noun a specific referable object.

Example: The robot is systematic.
In English the only form of the definite article is the, while in languages that mark for other properties there are more variant
e.g. Dutch you use de for the singular nouns (masculine singular, feminine singular), and het for neuter nouns.
e.g. German has six, marking noun gender and plurality

Keep in mind that the text as a whole may make indefinite references more definite later. For example, A robot is systematic. His name is Bob. is narrowing down, and in combination gives more information than 'The robot is systematic'.

A partitive article specifies an indefinite amount of a noun in its non-countable form (assuming it has one)

like the role du serves in French.
This does not strictly exist in English, though words like some serve a similar purpose.

A zero article is a hypothesized interpretation where the absence contrasts with the presence of an article, and analogous to a form the language may or may not have - for example, the collective of the respective noun

Example: Robots are systematic.
...or not so much and be marked, or ungrammatical, or whatnot
Example: Robot is systematic[1]

See also


Adpositions relate objects to something else; adpositional phrases are used to syntactically relate an object (often an NP) to something else, assisting someone's syntactic and/or semantic parsing.

The term 'adposition' is used as a convenient catch-all term to group:

  • prepositions, which are placed before (and the most common adposition in many languages)
  • postpositions, placed after
  • circumpositions, placed before and after.
  • ambiposition, may come before or after
  • inposition, placed inside


  • Adpositions are generally short. Prepositions and postpostitions are usually single words. Multi-word adpositions are regularly fixed sequences (and will show up as collocations).
  • Most languages have a fairly small amount of adpositions. They are taken to be a relatively closed lexical class.
  • languages that allow fairly free sentence reordering may allow prepositions to be moved, e.g. around the object.
  • Adpositional phrases may be an everyday part of a language's grammar. See e.g. the english examples.

In English

In My cat is on the sofa,

on is a preposition,
on the sofa is a prepositional phrase (verify).

Other examples of prepositions: astride, beside, by, despite, for, inside, like, of, on, out, till, throughout.

Some multi-word prepositions include in spite of, except for, next to.

English mostly uses prepositions, i.e. puts the adposition before what it describes.

There are a few postpositions,

such as in miles away, years on and that aside.
Various words used as postpositions are also used as prepositions
Some postpositions are arguably adverbs too, or instead.

And a few circumpositions, such as ' from then on.'


Adpositional phrases:

  • Adpositional phrases often act as an adjective or adverb.
    • Example: In "The book is on the floor", 'on the floor' is an adpositional phrase that acts as an adjective.
    • Example: "Before class, I threw the book", 'before class' acts as an adverb detailing time.
  • As dependent phrases, they never contain a sentence's main subject (interesting factoid to parsers; we tend to sense this, but this is not always obvious to parsers as they may easily appear before the main subject)


Words that do little more than express the speaker's emotion, often words or phrases, sometimes longer.

They tend to be syntactically/grammatically insignificant.

Includes empty words like an initial 'well,' arguably stopwords like 'hmm,' 'um' and 'er,' exclamations like 'cheers' and 'damn,' simple greetings like 'hello,' and also 'sorry,' and even 'excuse me.'

Open and closed classes

An open/closed class usually refers to properties of a lexical category / word class.

A class being closed refers to it being a static group, one that doesn't continually acquire words, or at best a lot more slowly.

In many languages, closed classes include determiners, prepositions (adpositions in general), pronouns, conjunctions, and particles.

An open word class is a category of words into which things can easily be borrowed, coined, compounded, derived, and so on. Nouns, verbs and adjectives are probably always open, and interjections often are as well.

Auxiliary verbs may have a common core that is closed, but is otherwise relatively open(verify).

Openness versus closedness regularly comes up in controlled corpora and computational linguistics. For example, WordNet does not contain certain small closed classes, apparently because they're expected to be known.


Substantive use refers to use of a word outside its most common lexical category.

'Book', for example, can refer to the paper thing you can read as well as the action of reserving a flight. There are various other nouns-as-verbs.

Adjectives may be used as nouns to refer to groups with the described property, such as in "The reds versus the blues."

Nouns may be used as adjective of sorts, particularly in the first part of a noun-noun compound.

'Be' is usually copulative, but in the existential 'to be' it is a 'verb substantive.'

In some ways, words that do this are concept words (not unlike a gloss or lemma) that contain a quality that can be geared to multiple uses.

A more extreme view lies in the idealisation of this concept, in which substantives are a separate lexical group that can be used as nouns, verbs or modifiers purely by their use in a sentence, and therefore describe objects, actions and qualities.

Isolating languages see some of this, but it seems that that particularly artificial language designers are fond of this idea. Relatively few languages are so analytically geared to make this true on a large scale, or true without having many idiosyncracies.

English is quite synthetic, easily leading to substantive uses that go on to live separate semantical lives. Consider the concept of chairing a comittee has only a tangential semantic relation to the chair the chair sits on.

Not all cases are so complex. Creative uses of nouns as verbs, such as "they didn't language that proposal very well," has a meaning that is naturally intuited as an active analogue to the original noun's sense.


Some people specifically mean substantive to mean 'anything usable as a noun' (perhaps the most common type of substantive, at least in English), which includes nouns, gerunds, adjectival nouns and pronouns.