Video 2.1. Parts of Speech

Summary

In this Syntax video tutorial, Professor Andrew Carney from the University of Arizona explores the intricacies of parts of speech, challenging traditional semantic definitions and emphasizing the importance of syntactic categories in sentence construction. He discusses the limitations of intuitive understanding and highlights the need to consider distributional properties, such as affixes and syntactic positions, to accurately identify parts of speech across languages. The video also touches on the role of function words and their grammatical significance beyond semantic meaning.

Takeaways

• 📚 The video is part of a series of tutorials on syntax, led by Professor Andrew Carney, focusing on parts of speech as the building blocks of sentences.
• 🧩 Parts of speech are crucial for syntacticians because they determine how words combine in sentences, affecting their positions and functions within those sentences.
• 🔍 Traditional definitions of parts of speech, such as nouns, verbs, adjectives, prepositions, and adverbs, are based on semantics but often fail to account for exceptions and cross-linguistic differences.
• 🎥 The video references 'Schoolhouse Rock' as an example of outdated and incorrect semantic definitions of parts of speech that were popular in the past.
• 🚫 Semantic definitions are problematic because they do not hold up when examining words with identical or closely related meanings that can have multiple parts of speech.
• 🌐 Cross-linguistic comparison shows that semantic definitions of parts of speech do not apply universally, as the same word meaning can have different parts of speech in different languages.
• 🔬 Syntacticians use distributional definitions for parts of speech, focusing on morphological and syntactic distribution, rather than semantic definitions.
• 🔠 Morphological distribution refers to the affixes attached to words, while syntactic distribution refers to the position of words in relation to others in a sentence.
• 💡 The video emphasizes that while the definitions of parts of speech based on distribution may seem circular, they are practically describing the relationship between a word's form and its position in a sentence.
• 📝 The script provides examples of the distributional criteria for different parts of speech in English, such as the use of articles before nouns and the position of adjectives and noun phrases.
• 🔑 The video concludes by highlighting the importance of understanding parts of speech not just through their meanings but through their distribution in language to accurately identify them.

Q & A

What is the main focus of the video 'Syntax: A Generative Introduction'?

The video focuses on the concept of parts of speech, which are of interest to syntacticians because they are the building blocks of sentences and determine how words can be combined.

Why are parts of speech important in syntax?

Parts of speech are important because they allow us to understand which items can appear in which positions within a sentence, and they help in identifying the syntactic categories that syntax manipulates.

What is the issue with the intuitive definitions of parts of speech based on semantics?

The intuitive definitions based on semantics are problematic because they do not account for edge cases where a word with a certain meaning can function as a different part of speech, and they do not hold true cross-linguistically.

Can you provide an example of a word that defies the semantic definition of parts of speech?

The word 'assassination' is an example. Semantically, it represents an action, but in the sentence 'The assassination of the president,' it functions as a noun.

What does the video suggest as an alternative method to determine parts of speech?

The video suggests using distributional definitions, which focus on the affixes attached to the word and the syntactic contexts in which the word appears, rather than the word's meaning.

What is the difference between morphological and syntactic distribution?

Morphological distribution refers to the affixes on a word, while syntactic distribution refers to the position of the word relative to nearby words in a sentence.

Why is the classic definition of an adverb considered 'terrifying' in the video?

The classic definition of an adverb is considered 'terrifying' because it is overly complex and includes a long list of qualities that adverbs can express, making it difficult to conceptualize as a natural class.

How does the video address the issue of words with multiple parts of speech?

The video points out that words with closely related meanings can have multiple parts of speech, such as 'father' being used as a noun, verb, and even an adjective, which challenges the semantic definition of parts of speech.

What is the role of function words in a sentence according to the video?

Function words do not have a specific meaning or reference; instead, they serve a grammatical purpose, helping to structure the sentence and connect its components.

How does the video explain the ability of native speakers to identify parts of speech in made-up words?

The video explains that native speakers can identify parts of speech in made-up words by looking at the affixes and syntactic positions of the words, rather than their meanings.

Outlines

00:00

📚 Introduction to Syntax and Parts of Speech

Andrew Carne introduces the concept of parts of speech in linguistics, emphasizing their importance as the building blocks of sentences. He explains that syntacticians study these categories to understand how words combine to form sentences. Carne critiques traditional definitions of parts of speech based on semantics, pointing out that they fail to account for edge cases and do not apply consistently across languages. He mentions the influence of 'Schoolhouse Rock' on common misconceptions and introduces the idea that parts of speech are syntactic categories, manipulated by syntax rather than defined by word meanings.

05:01

🔍 Challenges with Semantic Definitions of Parts of Speech

This paragraph delves into the inconsistencies of semantic definitions for parts of speech, using examples such as 'assassination' and 'sincerity' to illustrate how words can belong to different categories based on their use in a sentence, despite their inherent meanings. Carne discusses how words with related meanings can function as different parts of speech in various sentences, challenging the semantic approach to classification. He also highlights the cross-linguistic differences in parts of speech, using examples from Irish, Quamera, and Walbree to show that semantic definitions do not universally apply.

10:03

🌐 Beyond Semantics: The Role of Syntax in Parts of Speech

Carne argues that syntax, rather than semantics, is the key to identifying parts of speech. He demonstrates this with a sentence containing made-up words, showing that native English speakers can still categorize these words as adjectives, nouns, verbs, or adverbs based on their affixes and syntactic positions. This indicates that our understanding of parts of speech is rooted in their distribution, both morphologically through affixes and syntactically through their positions in sentences, rather than their meanings.

15:03

📝 Distributional Approach to Defining Parts of Speech

The paragraph explains the distributional approach to defining parts of speech, focusing on morphological and syntactic distribution. Morphological distribution refers to the types of affixes attached to a word, while syntactic distribution pertains to the word's position relative to other words. Carne clarifies that definitions of parts of speech are language-specific and provides examples of how nouns, verbs, adjectives, and adverbs are identified in English through their distributional properties.

20:05

🔚 Summary of Parts of Speech and Upcoming Discussion

In the concluding paragraph, Carne summarizes the importance of parts of speech as the foundational elements of sentence structure and the limitations of semantic definitions. He stresses the need to consider morphological and syntactic distribution for accurate classification. The paragraph ends with a preview of the next video, which will address functional parts of speech and methods for their identification.

Keywords

💡Parts of Speech

Parts of speech are the fundamental categories of words in a language, such as nouns, verbs, adjectives, adverbs, and prepositions. They are central to the video's theme as they are the building blocks of sentences and determine the syntactic structure. In the script, the professor discusses how traditional definitions of parts of speech based on semantics are problematic and instead emphasizes the importance of distributional properties in identifying them.

💡Syntax

Syntax refers to the set of rules in a language that defines the structure of sentences, specifically how words combine to form phrases, clauses, and complex sentences. The video tutorial focuses on syntax because it is through the understanding of parts of speech that syntacticians can analyze sentence structures. For example, the script mentions that syntax manipulates syntactic categories, such as parts of speech, to determine allowable positions for words in a sentence.

💡Semantics

Semantics is the study of meaning in language, which traditionally influences the classification of parts of speech based on the meanings of words. The script critiques semantic definitions of parts of speech, pointing out their limitations when it comes to edge cases and cross-linguistic consistency. It contrasts semantic definitions with distributional approaches in linguistic analysis.

💡Syntactic Categories

Syntactic categories are classes of words that are defined by their roles in sentence structure rather than their meanings. The script discusses how parts of speech are syntactic categories because they are manipulated by syntax to form grammatically correct sentences. The professor argues that these categories are better identified by their distribution in sentences rather than their semantics.

💡Distributional Definitions

Distributional definitions in linguistics refer to defining parts of speech based on where words appear in a sentence (syntactic distribution) and what affixes they take (morphological distribution). The script emphasizes that these definitions are more reliable than semantic ones for identifying parts of speech, as they are not dependent on the meaning of the words.

💡Morphological Distribution

Morphological distribution pertains to the types of affixes (prefixes, suffixes) that can be attached to a word, which can indicate its part of speech. The script uses morphological distribution as a criterion for identifying parts of speech, such as the '-ish' suffix suggesting an adjective or the '-er' suffix on nouns.

💡Syntactic Distribution

Syntactic distribution is the pattern of positions that words of a certain part of speech can occupy in a sentence. The script explains that understanding syntactic distribution is crucial for determining the part of speech of a word, as it is not always predictable from semantics alone, as shown by examples of words that can have different syntactic roles despite similar meanings.

💡Function Words

Function words are words that serve a grammatical purpose without necessarily having a concrete meaning or reference. They are used to structure sentences and include conjunctions, prepositions, and articles. The script mentions function words as a separate category that does not rely on semantics for their identification but rather on their role in sentence structure.

💡Intuitive Definitions

Intuitive definitions are those that are commonly understood or taught without rigorous linguistic analysis, often based on the semantic content of words. The script critiques intuitive definitions of parts of speech, such as 'a noun is a person, place, or thing,' for being overly simplistic and not accounting for the complexities of language structure.

💡Edge Cases

Edge cases in linguistics are examples that do not fit the general rules or definitions and thus challenge the conventional understanding. The script uses edge cases to illustrate the limitations of semantic-based definitions of parts of speech, such as the word 'assassination' functioning as a noun despite describing an action.

💡Cross-Linguistic

Cross-linguistic refers to the comparison of linguistic features across different languages. The script discusses the importance of cross-linguistic analysis in understanding that parts of speech are not universally defined by semantics but must be identified based on the specific syntactic and morphological properties of each language.

Highlights

Introduction to Syntax: A Generative Introduction, Fourth Edition by Andrew Carney, a linguistics professor at the University of Arizona.

Discussion on parts of speech as the building blocks of sentences, crucial for syntacticians to understand sentence construction.

Parts of speech, such as nouns, verbs, adjectives, prepositions, and adverbs, determine the possible positions of items in a sentence.

The traditional definitions of parts of speech are based on semantics and are often incorrect when scrutinized.

The limitations of semantic definitions are exposed through edge cases where words defy their traditional categories.

The Schoolhouse Rock video 'A Noun is a Person Place or Thing' is criticized for being outdated and incorrect.

Words with identical or closely related meanings can have multiple parts of speech, challenging semantic definitions.

Cross-linguistic comparison shows that semantic definitions of parts of speech do not hold universally.

Examples from Irish, Quamera, and Walbree demonstrate that semantic definitions fail to account for syntactic behavior.

The importance of syntactic position and affixes in determining a word's part of speech, independent of its meaning.

Syntacticians use distributional definitions for parts of speech, focusing on morphological and syntactic contexts.

Morphological distribution refers to the types of affixes attached to words, indicating their part of speech.

Syntactic distribution is about the position of words in relation to others, aiding in part of speech identification.

The circularity in the argumentation of defining parts of speech by their position and using position to identify them.

Examples of distributional criteria for nouns, verbs, adjectives, and adverbs in English.

The need to understand rather than memorize the criteria for identifying parts of speech.

Upcoming discussion on functional parts of speech and their identification methods in the next video.

Transcripts

[Music]

welcome to syntax a generative

introduction

fourth edition my name is andrew carney

i'm a professor of linguistics at the

university of arizona

i'm the author of your textbook and i'll

be leading you through this series of

video tutorials

in this chapter we're going to discuss

parts of speech

this is of interest to syntacticians

because

we want to know what the items are

that we're going to combine together

into our sentences

so we're going to look at what are

effectively the building

blocks of sentences these are things

like nouns

verbs adjectives prepositions and

adverbs

they're categories that allow us to tell

what items can appear

in which position we care about this

because

we find that some items can appear in

some positions and some items can appear

in others

but they don't always overlap so for

example the sun shines too brightly in

tucson

sun is a noun and it can appear as the

subject of the sentence

whereas a verb like will glow can't

appear in that position

we also call parts of speech syntactic

categories

because they're the categories that

syntax manipulates

now you might think it's strange for us

to cover this topic

because a lot of people have an

intuitive sense of what these parts of

speech are

so for example you may have heard of

such things as a noun as a person place

or thing

it's maybe even been drilled into you in

school

when i was a a kid back in the 70s and

80s

on saturday mornings we used to have

this television show called schoolhouse

rock

and in schoolhouse rock they had great

videos

of with animations and fantastic music

and they taught you important ideas

about grammar government and even

science

but unfortunately when it comes to

grammar they were way off track

they were based on very old ideas

about what parts of speech and other

grammatical categories are

um so for example you can

go online into youtube and and search

for schoolhouse rock

and you'll find a video called a noun is

a person place or thing and it's a

lovely video

but it's just deeply deeply deeply

incorrect uh i can't include it here in

this video because that would be a

violation of

uh copyright but there's a link there on

the screen

that you can use or you can search for

it

now these classic definitions these

intuitive definitions that you may have

learned before

have one thing in in common they are

based

on semantics they're based on the

meanings

of the words so for example a noun is a

person

place or thing so if you look at a word

like

um a doctor doctor is a person

therefore the word doctor is a noun uh

verbs are actions occurrences or states

of being

adjectives are modifiers that express

quality quantity or extent

so all of these are based in the meaning

of the word

that we're trying to define in terms of

part of speech

this next definition is terrifying if

you ever see a definition like this in

linguistics you better

grab your wallet and run adverbs are

modifiers that express

manner quality place time degree

number cause opposition affirmation or

denial

i don't know how you're supposed to

think of that as a natural class

prepositions are modifiers that indicate

location or origin

so as i said before all of these

definitions

are based on the semantics of the word

and for the most part

they correctly capture which part of

speech a word is

but the problem is that once you start

diving into this

with a little more rigor and detail you

discover that the

edge cases show us that semantics is not

really how we're figuring out what a

part of speech of a word is

so um let's take for example the fact

that

in many languages

you find words that meet one definition

but in fact are categories of another

let's take so

for example the word assassination by

any definition the word assassination is

an action

but clearly in the sentence the

assassination of the president

it's a noun similarly sincerity

because it describes equality is an is

suppose

should be an adjective but here in this

sentence sincerity is an important

quality

it's a noun to sort of really stretch

the example

our definition of a preposition

is indicating a location so take tucson

is where new yorkers flee for the winter

clearly tucson here is not a preposition

so uh when we look at these edge cases

we see that the definitions break down a

little bit

we also see this when we look at words

with

either identical or very closely related

meanings

that can have multiple parts of speech

so for example the word father

can be a verb it can be a noun and with

a little bit of a stretch it can even be

an adjective

the meanings are all related to each

other so you would expect that they

would

have the same part of speech if we

define parts of speech using semantics

but in the first sentence father is a

noun

second sentence father is a verb and

then the third one

it's an adjective

this is compounded when you start

comparing languages to each other

so if parts of speech are defined using

semantics

then you would expect that a word that

means the same thing

in two different languages so for

example

the word doctor in english and the word

doctor in irish

refer to doctors right so that's

semantically we know that they refer to

the same thing

you would expect them to always have the

same part of speech

but this clearly isn't true we're going

to look at three examples here one from

irish one from coimera and one from

walbree

that show that the english definitions

don't work cross-linguistically

so take irish here so in irish

verbs have a very specific position

irish is a verb subject object order

language

v-s-o and verbs always appear in a very

specific

privileged position they follow certain

kinds of particles

like negation they precede agreement

morphology

and they precede subjects

so look at that first sentence in a e

is the verb to eat and it follows the

negative

particle knee and it precedes both

the third singular masculine agreement

marker

and the subject john so niehan

now that's where verbs go now if we look

at the sentence in b

this is a bit of a surprise because here

we're finding a noun

in exactly that position so we're

finding

the word doctor between knee and e

it's a different third singular

agreement marker but it is in fact a

third singular agreement marker

so nidakdore that doctor is appearing

before the agreement and the subject

just where verbs do

and it appears after negation just where

verbs do

so doctor here seems to be functioning

like a verb

even though it's semantics tell us it's

a noun

let's look at quamera in quemera

verbs can be marked with certain

prefixes that

indicate the subject so for example

the subject yak is marks the first

singular the first singular means i

so if you want to say i don't like this

dog or i dislike this dog

you put yuck in front of the word for

dislike

now in the b sentence we should be

surprised to see

that the same prefix the one that means

i

is attaching itself to what semantically

is an adjective so the word for small

is uh has the property of expressing a

quality and it's taking this prefix

that only attaches to um to verbs

so that this tells us that

something else is going on here small

which by semantics should be an

adjective

is functioning like a verb the

in the quomaria example small was

functioning like a verb

in wallbury the word for small is

functioning like a noun

so uh in walbree which is a language

spoken in australia

the word for small can take this subject

marker

lu um which uh is always attached to

nouns

always attach the subject noun phrases

and here you'll see

that the the word small is appearing in

this position

the best english translation we can come

up with is something like the small one

but there's no noun uh one

in the wall breed instead what you have

is is you're just taking that subject

marker and sticking it straight on to

the word for small

so here again the semantics breaks down

it still is indicates the the quality of

being small

but it's it's functioning like a noun

if you're not convinced i think this

next um

case will really convince you take this

sentence

filled with made-up words of in uh for

english the incash dripner

into the ninden with epitips now what's

particularly interesting about this

sentence

is i sincerely hope that none of you

know what yinkish means

or drippner or

these are all made-up words so you

cannot know

what the semantics of them are you

cannot know what their reference

is nevertheless every native speaker of

english

will tell you that yankish is an

adjective that dripner is a noun

blorked is a verb questionably is an

adverb and both ninden and pittibs are

nouns

how did we know this if we don't know

what they mean

how do you know what griffner is if you

don't know

what a driftner is how can you tell if

it's a person

or a place or thing

the answer to this is we don't actually

use semantics to figure out parts of

speech

instead what we're looking at and

referring to

are the affixes that means the suffixes

and prefixes

and the syntactic position of the words

that we're looking at

so for example yinkish has on it

this ish suffix that's a typical suffix

for adjectives

driftner ends in er that's a common

suffix on nouns

we also note that dripner follows

yinkish which we've deduced to be an

adjective and the article

the this is a very typical position

for nouns to appear in blorked

is taking the ed suffix which we take to

be

a um a past tense marker and it's

following the subject

again this is a very typical position

for verbs

chrostophically has got that l y suffix

that tells us it's probably an adverb

nindin is immediately appearing after

uh an article that's a typical position

for nouns

pithibs that's true for pittibs too and

pittibs also has

this suffix s which um suggests to us

that whatever epididym is there's more

than one of them

so um we use these suffixes

and we use the syntactic position to

figure out what the part of speech is

even though we don't know what these

words mean so that tells us

that the semantic definitions a noun is

a person place or thing a verb is an

action occurrence or a state of being

those kinds of definitions

are while intuitive are almost certainly

wrong

so there's one more point i want to make

before we talk about

how we actually define parts of speech

which is

that there's a whole set of words which

don't really

have a meaning in the sense that they

don't have a reference

instead they have a function so they

have a grammatical purpose

but they don't actually have uh

something you can point to in the world

or identify in our imaginations

um we only use them to sticks uh the

bits of a sentence together

these are function words so take for

example that

of and for in the sentences on this

slide

you'll see that those underlined words

don't have a classic

meaning instead what they have is a

function

what they do is is uh stick the sentence

together so and i think that jose is in

the kitchen

that is um a complementizer

which functions to stick one clause

inside of another

and of is a preposition that links two

noun phrases together

okay so the what kinds of this

definitions are we actually going to use

to determine what a part of speech is

the definitions that syntacticians use

are distributional what we care about

is what affix is attached to the word

and what syntactic contexts

the word appears in

now it goes without saying because

language is different what affixes they

have

and how their syntaxes are structured

where these parts of speech definitions

are all going to be language

specific so the parts of speech

definitions for irish quemera

and walbury are going to be different

than those for

english okay so

we really have two kinds of distribution

we've looked at i've hinted at this

before

there's morphological distribution which

refers to

what the affixes on the word are and

there's syntactic distribution

which is the position of the word

relative to nearby words

i want to stop here for a second and ask

you to

think critically about one thing that we

have talked about

i started this discussion with the

observation that we

care about parts of speech because they

tell us

where in the sentence particular words

can appear

but now i've just defined parts of

speech

in terms of where the words appear

ask yourselves whether this is a

circular

fact or not have i done some circular

argumentation

by saying we need to know what the part

of speech something is by where it

appears

and then defining the part of speech by

where it appears

logically it's a bit of a circular

argument

but in fact in practicality it's really

just describing two sides

of the same fact that certain words

appear in certain positions

okay now we can list

the distributions for any particular

language

for particular parts of speech so for

example in english

nouns have uh particular morphological

distributions they take

particular kinds of derivational

suffixes these are suffixes

that turn words into nouns so things

like

um ship and ist and ness

and hood those are all suffixes that

indicate that the word we're looking at

is a noun similarly we have inflectional

suffixes like the plural suffix

s you have to be a little careful with

the plural s though

because s and its variants

also show up on verbs but interestingly

when it shows up on verbs it means the

exact opposite it means singular

means third person singular so for

example uh

dogs the the s the z or s that you hear

on dogs

um marks the plural but if i say

um he eats the s there

actually indicates that it's third

person singular

so just be a little careful with that

test we also can look at the syntax

so for example um articles appear before

nouns

adjectives appear before nouns

noun phrases follow prepositions

noun phrases can appear as the subjects

or direct objects of sentences

and if you're going to negate a noun as

opposed to negating a whole sentence you

use no instead of not

now it's possible to do this for um

any part of speech um we can do verbs

adjectives adverbs etc we can do lists

like this

i'll show you them very quickly if

you're actually interested

you could pause the video and look at

what's what's listed on each of the

slides

but you shouldn't try to memorize these

you should just understand

that these are the kinds of criteria

that we can use for determining parts of

speech

so here are some of the criteria we can

use for verbs

i won't go through these in too much

detail

similarly here are some criteria's the

criteria that we can use for adjectives

and here are some criteria that we can

use for adverbs

again don't memorize these don't try and

learn them

but understand how they function and how

they allow us to identify

something as a part of speech when not

relying

upon the meaning of the word

okay to summarize what we've talked

about in this video

is that we care about parts of speech

because they form the building blocks of

sentences

they tell us which words appear in which

positions

the classic definitions of these parts

of speech

while intuitively correct are

highly problematic they don't catch the

edge cases

they don't capture cross-linguistic

cases

and they don't allow us to understand

how we can

actually identify a part of speech

without knowing

the meaning of that sentence

those um classic definitions are all

based

in meaning linguistically instead what

we can do

is look at their distribution both in

terms of their morphological

distribution

and their syntactic distribution in the

next video

we're going to talk about functional

parts of speech and how we might

identify

[Music]

those