Lexicología del Inglés.

1
 Index.
CEX.
1. Introduction. Lexicology as theory of lexical component...............................................................3

2. Lexicology: Social, conceptualist and intrinsic views of the lexicon..............................................5

3. Public lexicon (lexis, dictionary) and individual lexicon (mental)...................................................7

4. The concept of “lexeme” and criteria for its identification: Polysemy and homonymy (polysemous
and homonymous words.)....................................................................................................... .........10

5. Paradigmatic relationships: Synonymy, antonymy, hyperonymy, (co)hyponymy.........................12

6. Syntagmatic relations. Semantic fields / dominions............................................................... .......14

7. Structure and organization of the (mental) lexicon........................................................................16

8. General structure of lexical entries. Lexeme’s content 1: Phonological information……………...18

9. Lexeme’s content 2: Syntactic information………………………………………………………........19

10. Lexeme’s content 3: Morphological information..........................................................................23

11. Other information associated to lexemes: Pragmatics, sociolinguistics......................................26

12. Other information: Etymology, orthographic and multimodal......................................................29

13. The problem of the nature and localization of lexical signification. Lexis and encyclopaedia.....31

14. Frege: Sense and reference. Intension and extension................................................................33

15. The Classical Theory: Necessary and sufficient conditions.........................................................35

16. Fuzzy categories. Rosch and the Prototype Theory....................................................................38

17. Componential analyses of lexical meaning I. Jackendoff’s Conceptual Semantics.....................41

18. Componential analyses of meaning II. Pustejovsky’s Generative Lexicon..................................44

19. Atomistic analyses of lexical meaning: Fodor and the postulates of meaning.............................46

20. SEM content in lexical entries......................................................................................................48

21. Semantic structure of lexemes which denote situations (events and states) I.............................49

22. Semantic structure of lexemes which denote situations (events and states) II............................51

2
 1. Introduction. Lexicology as theory of lexical component.
The first and one of the most important actions to do in order to study lexicology is to distinguish the
differences between “word” and “lexeme”. First of all, “word” is the common or vulgar way to refer to
lexemes. Next, the term “lexeme” is also a different concept as it ends with the suffix –eme, which
means “lexical idea”, as it also happens with other terms such as “phoneme” or “sememe”, all of them
meaning “abstract unit of something”. To sum up, being seen the differences between both terms, for
lexicology the only important elements are lexemes and abstract meanings.

Other important term that is frequent used in lexicology is “lexicon”. This term is basically a technical
way of saying “dictionary”, and the most relevant to this kind of study is the mental lexicon, which
refers to the dictionary that each person has in its mind, namely their vocabulary. Another element
present in the mental lexicon is the native’s intuition, as it affects the understanding of the meaning of
a word. Finally, lexical component is also present in the user of a language, as words represent it in
the way of speaking.

In the theoretical approach of lexicology, there is one important person: Noam Chomsky, mainly
because it was the first time that a theoretical approach appeared. He created a scientific method for
linguistics, making it a science too. He believed that human beings are born with a language faculty, a
system that is later developed to acquire a language, but this only happens with human beings,
making it a specific ability of humans.

On the other hand, he also stated that we do not have direct access to this faculty, but indirect
access, being this only possible through particular occurrences of language that a native speaker
produces spontaneously. Furthermore, with this method is also possible to have a better picture of
the native’s mind. All of this proved that we have competence and performance, being the latter the
occurrences and the part of language that we can perceive with senses, which then lead us to the
competence, which are the brain’s internal mechanisms of language. With this, it was stated that
language is a mental property that lays in human brains even when it is not being used, and that
competence is the permanent knowledge of a language.

All of this being cleared, the aim of lexicology is the speaker’s lexical competence. Lexicology is also
called lexical semantics, which is loosely defined as “the study of word meaning”. For us, lexicology is
the study of the speaker’s lexical competence, or the knowledge of lexis of a speaker.

Now, it is important to differentiate again two terms: Linguistic information and encyclopaedic
information. The line between both of them is very thin, being the linguistic information the one about
words through senses, also known as cognition, and the encyclopaedic information the one about the
multiple definitions about these. With this, it can be stated that we are capable of naming elements of
the sensitive world and applicating language on them. In fact, words are interrelated into a network of
words.

This part of language, the lexical component, is the beginning of the study of language, as it then
arrives into syntax, phonology and semantics.

Finally, there need to be differentiated two different studies of language: Lexicology and lexicography.
The main difference is that the first one is a descriptive discipline, meaning that it deals with the

3
description of language and lexemes, while the other is an explanatory one, which means that it is a
practical, applied discipline that deals with the compilation of dictionaries and its methodological
problems.

4
2. Lexicology: Social, conceptualist and intrinsic views of the lexicon.
In lexicology, there are three perspectives: The social viewpoint, the conceptualist viewpoint and the
intrinsic viewpoint. However, some specialists such as Murphy only contemplate two of them: The
lexicon is “out there” (in the language community – social viewpoint) vs. the lexicon is “in here” (in the
mind of a language user – conceptualist viewpoint). For the third one, the definition would be the
following: Lexical meaning is “in language” (a self-contained part of the linguistic system – intrinsic
viewpoint).

To define and understand the meaning of the lexicon, there are two definitions: 1. A collection of
meaningful linguistic expressions from which more complex linguistic expressions are built. However,
this creates different paradoxes: How are they more complex? Should the other be simple words?
Are compound simple words? What happens to phrases, which are part of syntax and semantics?
Are they not meaningful linguistic expressions? The solution to these paradoxes is that those
expressions can be composed by many words, but it has to have a single meaning for all the words of
the expressions, leading then to more complex expressions, which are phrases.

2. A collection of information about words and similar linguistic expressions in a language. Again, this
definition raises new questions and problems: What kind of information? What are the similar
linguistic expressions? However, in this case, there is not an exact solution, as with this definition the
lexicon is not very well defined. What can be taken clear from it is that the information presented can
be of two types: Linguistic and encyclopaedic, which have a non-well defined line of differences
between both of them.

Taking into account both definitions, it is cleared that the lexicon can refer to: 1. A dictionary.

2. The vocabulary of a language (lexis).

3. A particular language user’s knowledge of his / her own vocabulary (mental lexicon).

For the study of lexicology, only the two ones are relevant, being the objects of study of this science.

Now, having presented the three perspectives of the lexicon, it should be seen their theoretical
implications. The theoretical implications of the idea of the lexicon “out there” (social perspective),
which is a collection of anything and everything that is used as a word or a set expression by the
language community, are:

- It is a cultural artefact (an idealisation).

- We all aim to acquire such a lexicon of the community to be able to communicate successfully. We
are users of language.

- It is not made up of all the individual lexicon of the speakers, as the lexicon is out in society, not in
the speaker.

- All of this follows the functionalist perspective of Linguistics.

- It has a perspective / normative function, as it tells you how language should be used in a proper
and correct way.

5
- The role of an “authority” (academic professors like Randolph Quirk, “official” communicators, etc....)
in a language exists. In the British Isles and its colonies, it is the BBC. There is also one dictionary
considered as the vital, which is the OED (Oxford English Dictionary. It gives plenty information of
words, such as for example when they were born).

The theoretical implications of lexicon in the mind of language users (conceptualist perspective) are:

- Every speaker of the language has his / her own mental lexicon, not existing a common lexicon.

- Individual speakers’ lexicons are very different from each other.

- Even if mental lexicons exist in individual speakers’ minds, we do not want to investigate anyone’s
particular lexicon.

- Lexical meanings are concepts of the knowledge in the minds of the speakers.

- The stress is on universally acknowledged semantic entities (properties, relations....).

Finally, the theoretical implications of the lexicon within language itself (intrinsic perspective) are:

- It is not external to language.

- Lexical meaning is “sense” (philosophical conception).

- The focus is on an imagined “ideal” speaker of the language. It is a balance between the “out there”
(social perspective) and the “in here” (conceptualist perspective).

6
 3. Public lexicon (lexis, dictionary) and individual lexicon (mental).
Our dictionaries, standard desk dictionaries, include different types of information in lexical entries,
namely: Spelling, pronunciation, meaning(s) and word class. Also, to a different extent and on the
basis of their target audience, they may also include: etymology, morphologically related items,
synonyms and antonyms and usage / pragmatic information (examples of use or labels).

One question that usually raises when working with dictionaries is if they include all the words of a
language, the so-called lexis. The answer to this is negative, as it is impossible for a number of
reasons: 1. Dictionaries are aimed at specific target audiences (pocket size, learners, etc....). This
means that there is only interest in the area of the audience, making it impossible to contain all words.

2. Therefore, they usually avoid technical terms, jargon, dialectal uses, etc.... (i.e.: They represent
“standard language”). Only specified dictionaries include this kind of lexis.

3. Most importantly, languages are in a constant state of changing and new words enter the lexis of
the language all the time. These new words are called “neologisms”. On the other hand, words that
are too old that they are no longer used are called “archaic words”, and they appear with “Arc.” in
dictionaries.

In order to enter a language, neologisms follow the following processes, all explained by Murphy:

- Coining: A new word in invented from previously meaningless sounds. Sometimes, words are
invented to resemble in sound what they symbolize. These are called “onomatopoeias” (from the
Greek roots for “name” and “make”), and they include cases such as bang for the sound a gun makes
and cuckoo as the name of a bird that makes that sound. More rarely, words are just created out of
thin air, though this mostly happens for proper names, like in the cases of Kodak or Idaho.

- Derivation: Affixes (such as prefixes or suffixes) are added to a word to make a new word, such as
disobey from obey, resignation from resign or teacher from teach. This type of affixation changes the
word class (part of speech) and / or the meaning of the base word.

It is opposed to inflection, in which affixes are used to change a word’s grammatical subclass, like for
example the tense of a verb or the number of a noun. It is not a word-formation device, as it does not
create a new lexeme, but an instantiation of an existing lexeme.

- Compounding: Two existing words, are joined to make a new word that combines their meanings,
like in backlash or bluebell.

- Blending: Parts of two words are combined in a way that combines their meanings. For example,
“brunch” takes it “br” from “breakfast” and its “unch” from “lunch”, and “urinalysis” comes from “urine”
+ “analysis. These are sometimes called “portmanteau words”.

- Clipping: The new word is created by shortening an existing word. For example, “fax” comes from
“facsimile” or “flu” from “influenza”. Such cases may not at first involve the introduction of a new
meaning into the lexicon, but the original and clipped forms may gradually drift apart in meaning. For
instance, “fax” came to be used as a verb (I faxed her the contract), but “facsimile” does not have that
use.

7
- Borrowing: A word or phrase from one language is adopted by a different language. For example,
“chauvinism” was borrowed into English from French. Frequently, borrowed words undergo some
changes in form and meaning in the new language. In the case of “chauvinism”, the original French
“chauvinisme” has lost its final “e”, and its meaning has changed over time so that it is less likely in
English to refer to “patriotism” (particular loyalty to Napoleon) as it first did in French, and more likely
to refer to “sexism”.

- Calque (also loan translation): A concept is imported from another language and the name for it is
literally translated. For example, when French needed a word for “skyscraper”, the English word was
translated into French as “gratte-ciel” (“scrapes sky”). English, in turn, translated the French “marché
aux puces” (“market of the fleas”) into “flea market” when it needed a term for that kind of market.

- Acronyms: A new word is created from the initials of a phrase. These may be “true” acronyms,
which are pronounced as they are spelt, such as “laser” (from “light amplification by simulated
emission of radiation), or alphabetisms, which are pronounced as a series of letters, like for example
“DIY” for “do it yourself”.

- Backformation: A word is created by removing perceived affixes from an existing word. In this
case, the original word looks as if it was derived from a simpler word, but it wasn’t. For example,
“orientation” existed in English as a noun derived from the verb “orient”. But since we are used to
seeing verbs with “-ate” and related nouns that end in “-ation” (like “narrate / narration” and “evaluate /
evaluation”), some English speakers perceived “-ion” as the suffix on “orientation” and “back-formed”
the verb “orientate”.

Consequently: 1. There is not such a thing as the lexicon of a language (this is a cultural idealization).

2. Individual lexicons are necessarily different. But different in what?

Speakers’ mental lexicon can vary according to at least the following factors: Age, education, work
expertise, social status, geographical origin, personal experiences, etc.....

In order to be competent in the use of a word, a speaker needs to know: its pronunciation, its
meaning, its word class (though speakers need not to be familiar with the terms noun, verb, etc.....)
and the morphologically related items. On the other hand, they do not need to know necessarily: its
spelling (writing is a social construct taught in school), its etymology and the regional or social
varieties (related to usage); although educated speakers may certainly have this knowledge for some
or many items.

Moreover, the representation of crucial lexical information is different from dictionaries in:

A) Pronunciation: Dictionaries give one phonetic transcription, but speakers may also be familiar with
different pronunciations and accents.

B) Meanings: Dictionaries usually give simple definitions so that they can be understood by a large
audience, but speakers may not know what a word (exactly) means (that’s why we use dictionaries
after all!), or they may be able to provide a more detailed definition if they are experts in the relevant
field.

8
Now, how can we access the mental lexicon? There is one way: Since it is a mental entity, we rely in
psycholinguistic data: Slips of the tongue or speech errors, experiments and introspection (native
speaker intuition, etc....). Murphy calls this “semantic intuitions.

But, do lexicographers also use semantic intuitions to compile a dictionary? The answer is that they
may, but they most often rely on: 1. Experts or authorities (e.g.: How famous writers use a given
word.).

2. Etymology (to decide on an ambiguous word).

3. And today, computer corpora.

9
4. The concept of “lexeme” and criteria for its identification: Polysemy
and homonymy (polysemous and homonymous words).
The notion of “word” is ambiguous. It could be: 1. Word as a “word form” or lexical unit. Ex: There are
seven words in this sentence.

2. Word as “lexeme”. Ex: Colour and color are two spellings of the same word.

With this clarified, words can be identified under the following criteria: Orthography, Semantics,
Phonology and Grammar. However, there is a problem with these criteria, and it is that they are often
contradictory. That’s the reason why in Lexicology we prefer the notion of “lexeme”, which is defined
by Murphy as “a linguistic form (i.e.: A bit of speech and / or writing) that is conventionally associated
with a non-compositional meaning.

Now, it is necessary to understand what “conventionally” and “non-compositional meaning” refers to:

- Conventional: Association of form and meaning known to the entire linguistic community.

- Non-compositional meaning: The meaning of a lexeme is not made up of the meaning of its parts.
Ex: “Houses” is a word, but it is not a lexeme (2 lexemes: “House” and “-s”).

Moreover, apart from simple words, the definition given by Murphy of lexeme includes also:

- Bound morphemes (“-er” in “driver”).

- Complex words with an unpredictable meaning (“greenhouse”, for example).

- Set phrases and idioms. (Ex: “give up”, “pull somebody’s leg”).

Note that the mental lexicon may include compositional forms if they are frequently used.

So, a lexeme is identified on the basis of its meaning, but many word-forms are contiguous. For
example, considering the word-form “port”, which can mean either harbour or a kind of wine, there are
two possible analyses: 1. Two distinct lexemes (homonymy), which is the one done in terms of
meaning.

2. One lexeme with two senses (polysemy).

How do we decide? We should identify discrete (contrary of contiguous) senses with relevant tests,
which are the definition test, the contrast test and the Zeugma test. Also, each discrete sense would
correspond to one lexeme.

Consider the following example of a Zeugma test: *The ship arrived at the port and then the captain
drank one glass of it. If you find this sentence a bit awkward, that’s because we have activated both
senses at the same time, which is not possible.

More examples: *John and hid driving license expired yesterday,

*They took the door off its hinges and want through it.

10
*John declined a drink and Peter a verb.

Murphy also uses “etymology” as a criterion to identify lexemes. Etymology is frequently used by
lexicographers, but it is dangerous in Lexicography. With Zeugma, “port” is homonymous, but not in
etymology. The meaning of Port referring to wine comes from the Portuguese city “Oporto”, which
literally translates as “the port”, so etymology tell us that “port” is polysemous, but, do average
speakers know this? No, and as we study mental lexicons, we cannot consider elements that average
speakers wouldn’t know.

Polysemy is rampant in language. Sources of polysemy include: Metaphor, metonymy, conversion or

zero-derivation and grammaticalization or semantic bleaching. Note that polysemy should not be
confused with vagueness. A lexeme is vague if it’s general in its application, like for example, “friend”.

Lexemes may also acquire different traits of meaning in context. For example, for the lexeme “car”:

- The car needs washing (the bodywork).

- The car needs service (the engine).

- We can’t afford the car (its prize).

We call this contextual polysemy. Also, sometimes polysemy may be systematic, having three
different alternations: - Count / Mass alternations: It can be seen for example in the lexeme “lamb”,
which could refer to the animal or the food.

- Container / Containee alternations: It can be seen for example in the lexeme “bottle”, which could
refer to the recipient or the amount.

- Product / Producer alternations: It can be seen for example in the lexeme “newspaper”, which could
refer to the institution or the object.

We call this “regular producer”.

11
 5. Paradigmatic relationships: Synonymy, antonymy, hyperonymy,
(co)hyponymy.
Words in paradigmatic relations belong to the same word class and share some characteristics. We
are interested in semantic paradigms, which involve senses that share many semantic properties, but
differ in some. These paradigmatic relationships are:

- Synonymy: Words that mean the same as each other (=). However, it is rare for two words to have
exactly the same meaning / use. To know if two or more words are synonyms, we can use
substitutability tests. These tests prove that words are substitutable if there is no change in the
meaning of a sentence when one word is substituted for another. Ex: Person = Human (synonyms),
person ≠ man (not synonyms).

There are three types of synonymy: Absolute synonymy, near-synonymy and sense synonymy.

1. Absolute synonymy: Words are absolute synonyms if they are substitutable in any possible
context with no changes in denotation or other aspects of meaning (including connotation).
Candidates for perfect synonymy are, for example, technical names for things like plants, animals and
chemicals. Ex: Furze = Gorse = Whin, groundhog = woodchuck.

2. Near-synonymy: Far more common is for words’ senses to overlap and be near-synonyms (≈).
Ex: Fake ≈ False, obtain ≈ acquire, john ≈ lavatory room ≈ powder room.

3. Sense synonymy: Also, words may have many meanings, so the word could have different
synonyms for each sense. Ex: My tummy feels a bit funny (= peculiar, ≠ comical) whenever I eat fish.
/ Anna told a hilariously funny (= comical, ≠ peculiar) joke. Because it is polysemous, we talk about
sense polysemy.

- Hyponymy / Hyperonymy: + Hyponymy: “Is a type of” relationship. It has to be differentiated the
“subordinate term” (hyponym) and the superordinate term (hypernym). Ex: Cheddar / Cheese:
Cheddar is a type of cheese (correct). *Cheese is a type of cheddar (wrong). This means that cheese
is the hypernym, while cheddar is the hyponym. Also, they have a transitive relationship: If cheddar is
a type of cheese, and cheese is a type of food, the cheddar is a type of food.

+ Functional hyponymy: “Is used as” relationship. It produces sometimes odd results and does not
work correctly. Ex: A specimen cup is a type of cup, and a cup is a type of drinking vessel, then a
specimen cup is a type of drinking vessel (this is odd, and even though it is possible, it is wrong).

On the other hand, hyponymy can be useful for checking for verbs and their meanings. The way of
doing this is by making a “is a type of” relation between the verbs (X-ing “is a type of” Y-ing). Ex:
Marching ˂ Walking ˂ Moving; which means that marching is a type of walking and walking a type of
moving, so marching is a type of moving.

It can be done something similar to the previous one for adjectives, but instead of having a “is a type
of” relation, it is a “is a way of” relation (being X “is a way of” being Y). Ex: Nauseated ˂ Ill; which
means that being nauseated is a way of being ill.

12
With all of this, there is one conclusion: Hyponymy is just a semantic relationship (i.e., only
denotational properties matter, connotational properties do no enter in hyponymy relationships).

- Antonymy: It is an incompatibility relationship involving two words (two co-hyponyms that share the
same superordinate term). Ex: Jumping is a type of moving, but it is not a type of walking, and they
are at the same level (jumping and walking ˂ moving), so jumping and walking are antonyms.

In antonymy, there is binarity, being this the only type of relation that is particularly binary. The
morphological processes to form antonyms are the following ones, all including prefixes that denote
opposition in the original word: “a(s)-“, “in-“, “un-“ and “dis-“. Ex: As-symmetrical, in-frequent, un-
happy and dis-satisfied.

Being this presented, it is clear that antonymy is a relation between two lexemes that share all their
relevant properties except for one that causes them to be incompatible, which means, those that have
minimal differences. However, there are also triplets, like for example hot / cold / cool or black / white
/ grey. In these cases, we tend to find two of them perfect examples of antonymy (hot and cold, black
and white), while they are not so perfect antonyms if the third word is taken into account (hot and cool
or black and grey). Conclusion: There is a scale to denote which lexemes are more antonyms than
others, existing with it a scale of antonymy. This scale is created thanks to the speaker’s perception.

Now, there are three types of antonyms: 1. Contrary antonyms (tall / short).

2. Complementary antonyms (even / odd). In this category, there is also a subtype, which is the one
of gradable complementary antonyms (dishonest / honest; they are gradable, like in very honest, but
others are not, like odd, as very odd cannot be said).

3. Directional antonyms. They can be converse (parent / child) or reversive (tie / untie).

Also, there is another type of antonyms, the contronyms, who are words that have two meanings that
are opposite or nearly opposite. They are also known as Janus words, as Janus was an ancient
Roman god with two faces that looked in opposite directions. Examples of these kinds of words are:

+ Apology: 1. A statement of regret for an action. / 2. A defence or justification of a cause or doctrine.

+ Dust: 1. To remove dust from (to clean by dusting). / 2. To sprinkle with a powdery substance.

+ Rent: 1. To obtain occupancy or use (of another’s property) in return for a regular payment. / 2. To
grant temporary occupancy or use in return for a regular payment.

13
 6. Syntagmatic relations. Semantic fields / dominions.
- Semantic fields.

Theories that treat the lexicon as thesaurus-like (words arranged in semantic fields, not alphabetical
order) hold that lexical entries are cross-referenced with or linked to other lexical entries. Meanings
could be similar or different (cross-referenced).

Some proponents of thesaurus models favour large networks of words that constrain one another’s
senses. Meanings are mutually dependent.

Semantic Field Theory (Adrianne Lehrer): Semantic space = semantic field, the semantic place is a
universe of meanings where words inhabit. Box diagrams, they are divided in boxes.

Related concepts: Move. Walk, crawl, run, jump, jog, saunter, stroll, amble, march, sprint and hop. A
native speaker would know which words are related due to the native speaker’s intuition. They may
not know their meanings, but they could relate or organise them, making a network. Not native
speakers will use dictionaries for that.

Horizontal = Contrast / Vertical = Inclusion

Move

Crawl Walk Run Jump1

Saunter Stroll / Amble March Jog Sprint Hop Jump2

Meanings: - Move: To change in position from one point to another.

- Walk: To move over a surface by taking steps with the feet at a place slower than a run.

- Crawl: To move slowly on the hands and the knees or by dragging the body along the ground.

- Run: To move using your legs, going faster than when you walk.

- Jog: To move with a jolting rhythm / To run or ride at a steady slow trot.

- Saunter: To walk at a leisurely pace / Stroll.

- Stroll: To go for a walk.

- Jump: To move suddenly an in one motion. / To spring off the ground or other base by a muscular
effort of the legs and the feet.

- Amble: To walk slowly or leisurely / Stroll.

- March: To walk steadily and rhythmically forward in step with others.

- Sprint: To move rapidly or at top speed for a brief period, as in running or swimming.

- Hop: To jump on one foot.

14
Semantic fields are established on the basis of: Hyponymy, co-hyponymy, contrast, synonymy and
partial synonymy.

Lexical gaps are also found, since not all languages have words for every box in the semantic space.
Mind too, that different languages may cut up the semantic space in different ways (setting different
boxes). Thus the difficult of translating certain words from one language to another.

In the case of colour words, such as red, purple, yellow, blue, lavender, scarlet, orange, green and
crimson, they should be divided in areas divided of nanometres, as specialists decide their meaning
with them, and also the mixture of colours. Because of this, their definitions are more difficult that
other word’s.

- Syntagmatic relations.

Syntagmatic relations are relationships between words that go together in syntactic phrases. Notice
that “syntagmatic” and “syntax” are from the same Greek roots, meaning “touching together”. They
are studied more and more these days as corpus research highlights they ways in which words tend
to occur with certain words rather than others. Ex: Asleep – Fast asleep, sound asleep, completely
asleep. But not “very asleep”.

These are called collocations. Halliday (1966) defines them as the tendency for words to occur
together repeatedly. Ex: Strong is similar to powerful. You can say both for arguments, but teas are
strong and not powerful, and cars are powerful but not strong. In these cases, it is not a problem of
grammar, but a problem of semantics.

Individual words may have collocationally restricted meanings. Consider: White wine (Yellow), white
coffee (Brown or with milk) or white man (Pinkish skin colour).

15
 7. Structure and organization of the (mental) lexicon.
Linguistics model of the lexicon: They assume that the lexicon is organized on the basis of lexical
relations: Synonymy, antonymy, class-inclusion (hyponymy), part-whole (meronymy), semantic fields,
etc.... But note that we are interested in the mental lexicon (ML) so, do those lexical relations have
psychological validity? Could there be a different kind of organization in the ML? How can we know?

In the study of the ML, we assume that:

1. Words in the ML must be organised in some way. Why?

A) It is believed than an average speaker stores about 40.000 words in their ML.

B) We retrieve and use lexical items at a rate of 2 – 3 per second.

So, if words in the ML weren’t organised some way, this would be impossible. So, how are words
organised in the ML then?

- Hypothesis 1: They are listed alphabetically just like a desk dictionary. But this can’t be true! It is
not psychologically correct, because illiterate people use lexical items efficiently, and speakers of
languages without writing systems use words as efficiently. So, we have to formulate a different
hypothesis.

We can use the following evidence for this: Speech errors, speech disorders and psycholinguistic
experiments.

A) Speech errors or slips of the tongue: They show normal lexical behaviour. They are made by
both educated and uneducated speakers. There are two types of speech errors:

A.1) Semantic errors (word substitutions): I wonder who invented crosswords (jigsaws) / He came
tomorrow (yesterday).

A.2) Sound errors: The emperor had two porcupines (concubines) / There were lots of little orgasms
(organisms).

B) Speech disorders: Aphasia: Speech impairment due to stroke, injury, etc.... Examples of word-
finding problems in aphasic patients: Chair for table, knee for elbow, hair for comb, tuber for butter or
leasing for ceiling. Can you identify any lexical relations? Yes, there is a semantic and phonetic
relation between them

C) Psycholinguistics experiments: We put the speakers to test. A trivial example of word

association: “Give me the 1st word that comes to your mind when I say X”. The strongest links are
usually: Co-hyponyms: Black / White, salt / pepper; hyperonyms: Red / Colour; antonyms: Big / Small,
low / high; and also collocations (salt water) or simply frequent combinations.

So this evidence ( A, B and C) together tell us: Words are related in the ML phonologically (sound
errors), semantically (through lexical relation like hyponymy, collocations, meronymy, etc....) and
syntactically: In a corpus containing 200 speech errors, Fay and Cutler showed that 99% of the cases
the intrusive word was of the same syntactic category as the target word.

16
Conclusion: The ML is a complex network of relations among lexical items. The ML has two modules:
A phonological one and a semantico-syntactic one. Thus, the ML is not a list of unrelated lexical
entries as in a desk dictionary (alphabetically), but it is rather a big and complex network of words.

Aphasias (Page 15 of the book).

There is research that shows you may lose your vowels with aphasia. Normally, the only vowel that
remains is the “i”, one of the most important and basic in the language.

Broca’s area: This area is important for language production. It is located in the frontal lobe. It
combines sounds into words. If damaged, sentences won’t be fluid.

Wernicke’s area: This area is important for language comprehension or interpretation. It is located in
the temporal lobe. It is used for creating speaking coherent sentences and for understanding speech.
If damaged, one cannot understand what they’re saying to him. People with aphasia in this area
normally don’t understand why people do not comprehend them, they do not perceive that they are
producing non-sensical or bad connected sentences.

Grammatical (function)

Words With aphasia, you lose one of them.

Lexical (content)

Types of aphasia:

- Fluent: People cannot understand other people nor produce their own speech.

- Non- fluent: People can understand other people but nor produce their own speech.

- PPA (Primary Progressive Aphasia): This is a kind of dementia which is first appearance is
related with language. The speech production starts to decrease gradually when this happens.

All of this proves that words are connected into language networks. There are two type of errors
produced with aphasia:

- Semantic: This happens when words are changed with synonyms or related words. Ex: Saying cat
instead of dog.

- Phonological: People say one word that is phonologically related to the one that they truly wanted
to say. Ex: Saying house /haᶷs/ instead of horse /hᴐ:s/.

Brain’s plasticity: It is a human ability in which our brain, after damaged, tries to be regenerated. This
has to be through therapy and linguistic exercises, as they can help to make the Broca’s and
Wernicke’s areas in the brain.

17
 8. General structure of lexical entries. Lexeme’s content 1:
Phonological information.
Not all the information of a word is available to native speakers necessarily. For example, we have
seen that etymological information is not available for most speakers. Similarly, the general
information we have on lexical concepts may vary.

A competent speaker should have at least pronunciation, definitions, syntactic context and related
inflectional forms for each lexical term of their language.

There are many different ways to state lexical knowledge, but we’d like our system of representation
to be:

1. Explicit, simple and faithful to the organization of the Mental Lexicon.

2. And, as an “additional bonus”, computer tractable.

This is what we find in the formalism known as: [ATTRIBUTE : VALUE] [A : V]. The reasoning behind
it is very simple: We define a number of attributes for a lexical entry as an initial template, and we
then assign a value for the lexeme at issue. Let us see an example:

The lexeme “organise” will contain the following phonological information: [PHON : ‘ ᵓ:gᵊnaiz]

However, we could be more specific in the PHON box and create further [A : V] pairs within it:

PHON: [SEGMENTS: /ssss…./]

[STYLL STR.: Syl (1) + syl (2)]

[STRESS: Syl (1)]

This complex combination of [A : V] pairs is known as a Feature Structure. In “organise:

PHON: [SEGMENTS: /‘ᵓ:gᵊnaiz/]

[SYLL STR.: [syl (1): ᵓ:]

[syl (2): gᵊn]

[syl 3: aiz]

[STRESS: syl(1)]

Note that we need not repeat the value ‘ᵓ: for “syl (1)” in “Stress”.

For example, for practical reasons, we will not pay much attention to suprasegmental phonology. So,
usually, the PHON box will be reduced to segmental phonology. [PHON: ‘ᵓ:gᵊnaiz ]. For the moment,
we will only use the DEF(inition) attribute. Its value will be a simplified dictionary-like definition. Ex:

[PHON: ‘ᵓ:gᵊnaiz] / [SEM: [DEF: “arrange things efficiently”]].

18
 9. Lexeme’s content 2: Syntactic information.
It contains two main feature structures: CAT, which refers to “CAT”egorial values: and C-
SELECTION, which refers to the number and type of complements the lexeme selects. Ex:

The water boils: 1 complement (S) (Necessary that something boils to make the action of boiling – 1
complement).

He likes music: 2 complements (S / O). (To like something, we need something and someone – 2
complements).

He gave me a book: 3 complements (S / O 1 / O2) (To give something to someone, you need the
object and the movement of the object through A to B – 3 complements).

In these cases, complements are equal as participants, which means that they are the obligatory
participants that is necessary to understand or interpret the meaning by the word that is being
analysed.

In CAT, MAJOR: Refers to the main word classes (V, N, A, etc….). Note that we include here open
and closed word classes (read pages 14 – 15 for close and open word classes). Categorial features
are linked to the MAJOR value. For example, GENDER is only applicable if MAJOR: N. In English,
only pronouns have grammatical gender.

AUXiliary is only applicable if MAJOR: V. Only some verbs can function as Auxiliaries,

COUNTable is only applicable if MAJOR: Nm but NUMBER and PERSON apply to both N and V.

Ex: Organise: [PHON: /‘ᵓ:gᵊnaiz/]

[SEM: [DEF: “Arrange things efficiently”]]

SYNTAX: CAT: [MAJOR: V]

[AUX: -]

However, because it is only an infinitive, there is no NUMBER or PERSON, but in the case it was
another form like “organises”:

[PHON: /‘ᵓ:gᵊnaiziz/]

[SEM: [DEF: “Arrange things efficiently”]]

SYNTAX: CAT: [MAJOR: V]

[AUX: -]

[NUMBER: Sing]

[PERSON: 3]

19
Another example of a different word, like Pronouns, could be the one of “he”, which acts as a N,
having with it GENDER and COUNT.

[PHON: /hi:/]

[SEM: [DEF: “Speech participant other than S and A]]

SYNTAX: CAT: [MAJOR: N]

[NUMBER: Sing]

[PERSON: 3]

[COUNT: -]

[GENDER: Masc.]

However, GENDER is only relevant in the Pronominal System.

Let us now consider the C-SELECTION attribute. As said earlier, lexical items can take complements.
Verbs at least one complement the Subjects (Water boils, John walks…). But many also take an
Object (The man is riding a horse, the water flooded the village…..) and some can take two Objects
(He gave me a book, my brother explained the problem to me.). Note that other categories can take
complements too:

Prepositions: In the kitchen, on the table, at the gate, the invention of the automobile, the king of
England, responsible for the situation, similar to mine.

Important: With the preposition “in”, we have to take into account how to use it correctly depending on
the phrases, being with it transitive or intransitive (she walked into the room - Also, bear in mind that
“after” is always a transitive preposition, and that for using it when talking about something that
happens later is necessary to use the adverb “afterwards”.

Let us continue with our example “organize”: This verb takes 2 complements (e.g.: The president
organized the World Summit). The subject is realized as a Noun Phrase (NP), and the object is
realised as a Noun Phrase (NP), which needs to be specified in the scheme of the word in the C-
SELECT part:

20
Organises: [PHON: /’ᵓ:gᵊnaIz/]

[SEM: [DEF: “Arrange things efficiently”]]

SYNTAX: CAT: [MAJOR: V]

[AUX: -]

[NUMBER: Sing]

[PERSON: 3]

C-SELECT: [COMPL (1): NP]

[COMPL (2): NP]

Consider now the entry for Locative “in”:

In: [PHON: /In/]

[SEM: [DEF: “Inside”]]

SYNTAX: CAT: [MAJOR: P]

[C-SELECT: [COMPL (1): NP]]

Or the noun “king”:

King: [PHON: /kIᵑ/]

[SEM: [DEF: “Male person who rules a nation”.]]

SYNTAX: CAT: [MAJOR: N]

[NUMBER: Sing]

[PERSON: 3]

[COUNT: +]

C-SELECT: [COMPL (1): PP “of”]

Note that the same item may select different types of complements:

1. John wants to come (O = Infinitival clause)

John wants a coffee (O = NP).

*John wants that Peter comes (* O = “that” clause)

21
C-SELECT: [COMPL (1): NP]

[COMPL (2): NP / Clause “to”]

The slash “/” indicates alternative values..

Note that the same item may select a different number of complements:

1. John is eating an apple.

2. John is eating.

C-SELECT: [COMPL (1): NP]

[COMPL (2): (NP)]

Parenthesis “()” indicates optionality.

22
 10. Lexeme’s content 3: Morphological information.
[M-TYPE: Idiom, word, stem, affix...]

Idioms are words, phrases or expressions that cannot be taken literally. In other words, the meaning
of an idiom is not derivable from the meanings of the containing words in combination.

Ex: “Break a leg”: Break a leg can be a free phrase (literal meaning) or an idiom meaning “do your
best and do well” (often, actors tell each other to “break a leg” before they go out on stage to
perform).

A word is a morpheme that can stand on its own, i.e., a free morpheme. It is preceded and followed
by blank spaces in writing. It is a concept and controversial concept.

Root / Stem: A root is a form that underlies at least one paradigm or partial paradigm, and is itself
morphologically simple / A stem is a form that underlies at least one paradigm or partial paradigm,
and is itself morphologically complex.

Ex: Divert / Diversion: Diversion: Stem (= Morphologically complex) / Divert: Root (= Morphologically
simple). A stem diversion underlies the singular and plural forms of diversion, and a root divert, the
varying forms of divert.

An affix is a bound morpheme that is joined before, after or within a root or stem. Affixes can never
stand on their own. An affix can be joined by derivation or inflection. Affixes can be:

1. Prefixes: Occurs before the root / stem. Ex: Unhappy.

2. Suffixes: Occurs after the root / stem. Ex: Happiness.

3. Infixes: Occurs inside of a root /stem. Ex: Fandamntastic. / Edumacation.

[M-CAT: V, N, .... (Esp. for affixes and roots)]

Word classes can be divided into open classes and closed classes. Open classes are readily open to
new words, closed classes are limited classes that rarely admit new words. These are:

- Open classes: Noun, adjectives, main verbs and adverb[s.

- Close classes: Pronouns, determiners, auxiliary verbs, conjunctions, prepositions……

Example of roots and affixes in different categories:

Affix root

N -ness table

A -ful big

V -ize rain

Adv -wise hard

23
Categories. Level: Word vs. level: Affix.

- A (Adjective): Level: Word = Big, ripe, nice, solid, strong, wise, profound.

- Aaf (Adjective affix): Level: Affix = -ful [N+_ ] (beautiful), -al [N+_ ] (cultural), -ous [N+_ ]
(dangerous), -ary [N+_ ] (evolutionary), -ly [N+_ ] (friendly).

- N (Noun): Level: Word = Beauty, culture, danger, evolution category, fragment.

- Naf (Noun affix): Level: Affix = -ity [A+_ ] (profundity), -ee [V + _ ] (employee), -al [V + _ ] (arrival).

- V (Verb): Level: Word = Employ, arrive.

- Vaf (Verb affix): Level: Affix: -ize [N + _ ] (categorize), -en [A + _ ] (ripen), -ify [A + _ ] (solidify)

- Adv (Adverb): Level: Word = Then, always, now.

- Advaf (Adverb affix): Level: Affix = -ly [A + _ ] (quickly), -wise [N + _] (weatherwise).

Unpredictable meanings:

- Categorize: “To put X into a category”. / “to convert or be converted into a category” / “To place in a category
(for treatment or care)”.

- Vaporize: “To convert or be converted into vapour” / “To put X into vapour” / “to place in vapour (for treatment
or care)”.

- Hospitalize: “To place in a hospital (for treatment or care” / “To put X into a hospital” / “To convert or be
converted into a hospital”.

When inserting the suffix “-ise” in a word, there are multiple meanings for it as affixes have their own
meanings. However, only one can make sense depending on the root of the word (the ones that are underlined
in the examples). The same happens with the rest of affixes.

[M-SELECTION: V, N, .....(esp. For affixes)]

The affix selects the category:

“-ness” selects the A category (happiness).

“-ous” selects the N category (dangerous).

“Un-“ selects the V category (unwrap).

M-STR.: [HEAD: [LEX:...]]

[COMPL: [LEX: ....]]

Morphological structure: What is the Head and what is the Complement? The head will be always what
determines the type of word. This can be seen in the analyses like the ones made in Introduction to English
Linguistics.

Ex: Happiness: [Happi] is the complement and [ness] the head.

Asleep: [A] is the Head and [sleep] is the Complement.

24
Inactive: [Act] is the complement and [ive] is the Head, next [in] is the complement and [active] is the Head.

INFLECTION: [PAST: -ed / allomorphs]

[PPT: -en / allomorphs]

[PL: -(e)s / allomorphs]

[COMPAR: -er / more]

[SUPERL: -est / the most]

An allomorph is an alternative morph for a given morpheme. Ex: [PAST: -ed / -t, 0, vowel change, phonological
change] (Switched / learnt, cut, drank, read....)

[PPT: -en / -ed, -t, 0, vowel change, phonological change…] (spoken / closed,, learnt, come, sat, read......)

[PL.: -(e)s / -en, vowel change, 0….] (tables / brethren, geese, sheep….)

DERIVATION: [LEX: .....]

[LEX: .....]

25
 11. Other information associated to lexemes: Pragmatics,
sociolinguistic.
When we open a dictionary and look carefully at the entries, we see that quite a number of them are
marked with labels if one kind or another. Ex: Botanical, baseball, slang, American English. These
labels indicate that the word is restrictive in its use.

In the “General Explanation” at the beginning of the first edition of the OED (1913), the editors
discuss the nature of the English vocabulary that the dictionaries intend to chart and describe,
offering the following diagram:

Scientific Foreign

Literary

Common

Technical Colloquial Dialectal

Slang

The OED recognizes a fundamental distinction: Common core words vs. words belonging to
particular specialist sub-sets. This distinction may apply not only between words, but also between
the senses of a single word. While some senses may belong to the common core, one or more sense
may be a part of a specialist vocabulary.

Dimensions of variation of vocabulary:

1. History.

2. Geography.

3. Occupation.

4. Association with social / cultural groups.

5. Formality of the context /(style).

6. Medium of communication.

1. Historical dimension: - Charting the birth and death of words.

- Words that continue to be recorded even if they are “obsolete”.

- Words, still in use, that are “archaic” (= an old-fashioned flavour, probably in the process of
disappearing).

26
2. Geographical dimension: - Represented in the OED diagram by “dialect”.

- Regional dialects of a single country,

- National varieties of English (as written / spoken in the US, Canada, Australia, India, West Africa...)

- Reinterpretation of the common core as the vocabulary of “international English” (=the words that
are common to all national varieties of the language).

- National varieties also have their common core.

Ex: “Book” belongs to the common core of all national varieties of English. / “Faucet” belongs to the
common core of American English and “tap” is the equivalent word in British English and other
Commonwealth varieties (Canadian English has both terms).

British English and American English: - The British and American varieties of English accounts for
around 70% of mother-tongue English speakers, with Americans outnumbering British by 4 to 1.

- American English is the dominant variety (political cultural and economic dominance of the US).

- Some words reflect cultural differences, with no equivalent in the other variety. Ex: American
English: Ivy League and Groundhog Day. / British English: Honours Degree and Value Added Tax.

- Some words are variety-specific but have an equivalent in the other variety. Ex: American English:
Baggage room and potato chips. / British English: Left-luggage office and crisps.

- Some American English words eventually became World English. Ex: American English + World
English: French fries, can and intermission. / British English: Crisps, tin and interval.

- Some British English words eventually became World English. Ex: American English: Casket,
drapes and line. / British English + World English: Coffin, curtains and queue.

- In some cases, both the British English words and American English words are part of World
English. Ex: American English + World English: Baggage and mail. / British English + World English:
Luggage and post.

3. Occupational dimension: Jargon: Specialist vocabularies associated with “occupations” that

people engage in (employment, leisure....). Types:

- Occupational jargons: Medicine, Computing, Financing....

Medicine: jargon based on Latin and Greek, especially in the formation of neo-classical compounds
(psychology, mammograph, audiogram...).

Computing: Novel compounds formed from established English words (touch sensitive screen, word
processor...), or new meanings for ordinary words (window, mouse, icon....).

- Religious language: Influence of Christianity on language (Saxon origin: Ghost, gospel, heaven,
worship....)

27
- “Green” jargon: Quite recent in time (acid rain, alternative energy...)

- Jargon of groups of people: 1. Youth culture: Awesome, dude, nerd, gangsta (music),
warehouse (party)....

2. Underworld slang: Cleanskin, go straight, a star....

3. CB (Citizens’ Band Radio) talk: Affirmative, doughnuts, grandma lane....

Dictionaries tend to label words that are towards the extremes of the styles:

1. Formal: Hereunder, carnivore, decease, ameliorate.....

2. Informal / colloquial: Booze, comfy, cuppa...

3. Slang / 4. Taboo: Dictionaries fluctuate in labelling some of these words. For a dictionary a word
may be informal, but for the next, slang. A word may be tagged as slang whereas for another is
taboo.

Basic distinction: Spoken / Written medium. There may be some words that we associate more
readily with either the spoken or the written medium.

28
 12. Other information: Etymological, orthographic and multimodal.
Speakers may associate further types of information to lexemes. These include.

1. Etymological information: This refers to the historical origin of lexical items, which may allow
speakers to deduce meanings and other linguistically relevant information.

Ex: Many medical terms are constructed with Greek morphemes. Consider the word “haematuria”:
Greek: “Haimatos” = Blood and Greek: “Urina” = “Urine”, which supposes that haematuria means
“blood in urine”.

So, etymology has helped to arrive at the meaning of the item, which is nice, but it is possible for
someone to know the meaning of “haematuria” without being aware of its Greek origin. So, speakers
need not to know the etymology of a lexical item to use it competently. However, speakers may be
unconsciously aware of etymological information. It is well-known in morphological theory that some
affixes typically attach to Latinate bases only.

Ex: signify, identity, investigate, federal. (Both are Latinate terms, the root and the suffixes)

2. Orthographic information: Educated speakers can also provide orthographic representations of

lexical items. Again, we have to remember that we study lexical competence as part of linguistic
competence.

Writing systems are secondary in languages (i.e., speech always came first) and indeed, many
languages in the world do not have writing systems, yet their speakers show the same degree of
fluency / ability in their linguistic behaviour.

Orthographic conventions are more related to linguistic performance than to linguistic competence,
but this is not the end of the story, though:

A) It is known that orthographic knowledge may speed up word-retrieval and recognition.

B) The Mental Lexicon has a module connecting words through their spelling similarities.

3. Multimodal information: Lexemes may evoke representations of different kinds: Acoustic, visual,
olfactory, etc.... For example, if I give you a word like “beer”, it is most likely that the typical
representation of “beer” appears in your mind, but it won’t appear: Beer - An alcoholic beverage
usually made form malted cereal grains (as barley), flavoured with hops and brewed by slow
fermentation (a definition which you can also produce, but only after some reflection).

These facts have led some to propose a theory of senses as “images”. This is called the Image
Theory of meaning (pg. 37). However, there are obvious problems with this theory:

A) There cannot be images for many lexemes of the language (i.e., abstract entities like “philosophy”,
“illness”, “time”, etc....

B) Speakers are unlikely to share similar visual representations for most items, which makes it difficult
to explain how we can communicate at all.

29
Nevertheless, it is obvious that visual representations exist in the mind and cognitive scientists have
proposed complex models to account for the relation between language and vision. That’s why we
have included an [IMAGE] attribute in our entries. However, for practical reasons, we will not make
use of it (we cannot append images / smells, etc.... to our entries), but you should remember that this
information is part of lexical competence.

30
13. The problem of the nature and localization of lexical signification.
Lexis and encyclopaedia.
We have to delimit what aspects of meaning count as “lexical semantics” and what aspects do not.
We will have to distinguish semantic kinds of meaning: Denotative [DEN] vs. Pragmatic, connotative
meaning or social meaning (different ways to refer to the same part).

As an exercise, think about what shoe describes in: Cinderella was going to be late for the ball
because she couldn’t find her other shoe. With this scenario, you can imagine:

1. The shoe is probably of a style appropriate to a woman.

2. A shoe that she intends to wear to the ball.

3. A dressy shoe.

4. (If you know the story of Cinderella) The shoe is a glass slipper.

All of them include information provided by the context. In other words, they are inferences we have
made based on the information available to us in the context. Such inferences are part of the realm of
Pragmatics (the study of context-bound meaning). But what counts as context? Two things:

1. The rest of the linguistic material surrounding the word (co-text).

2. Background knowledge (also called encyclopaedic or world knowledge).

So, change the context and you may interpret shoe differently. As in: “The mare was almost ready to
pull the carriage, but Cinderella was going to be late for the ball because she couldn’t find her other
shoe. She had three of them piled on the table in the blacksmith’s shop, but where was that fourth
one?”

New context: - Mare.

- Three shoes looking for the fourth one.

- At a blacksmith’s.

So now, shoe may refer to a horseshoe, not to Cinderella’s slipper.

Semanticists are: - NOT interested in pragmatic inferences.

- Interested in semantic entailments.

An entailment is a relation between propositions: if A then B. They identify necessary properties of

meaning. Examples:

1. If it is a shoe, then it is made to be worn on a foot (YES ).

2. If it is a shoe, then it is made for humans to wear (NO ).

3. If it is a shoe, then it is made of leather. (NO :v).

31
4. If it is a shoe, then it is size 10 (NO >:( ).

So, if something is a shoe, it is necessarily footwear, but it may also have other properties that are
incidental, rather than necessary (entailed) properties of shoe-ness. Incidental properties are
defeasible (they can be destroyed).

Note that dictionary entries include both necessary and incidental aspects of meaning.

Denotative (conceptual or cognitive) meaning: The relation between a word and the things /
properties / actions / concepts that it refers to. A word’s denotative meaning is its literal meaning
(dictionary meaning).

Connotative meaning: Any semantic association that a word has, which are not strictly part of the
denotative meaning of the word. Consider: Feline, cat, kitty. The three words can denote the same
things, yet they are different in their connotations:

- Feline <scientific>

- Cat <standard, neutral>

- Kitty <affectionate / diminutive>

Example: Snake.

- Denotative meaning: “Any of numerous scaly, legless, sometimes venomous reptiles having a long,
tapering, cylindrical body and found in most tropical and temperate regions”.

- Connotative meaning: Evil

Hope.

Denotative meaning: “Residence or fixed dwelling place”.

Connotative meaning: Privacy, coziness, intimacy, refuge, resting place.

Social meaning: Information about the person who is using a given word and the social situation
they are in. Ex: Howdy <American, rural are, informal, friendly towards the addressee>. A particular
type of social meaning is the one that denotes the connotations that society have of words. Ex:
Consider the words “drunk” and “alcoholic”. The first one shows disapproval, while the other one is
more neutral.

32
 14. Frege: Sense and reference. Intension and extension.
We are interested in denotative meaning. But what is denotative meaning made of? There are
different options:

1. Meanings are the referents of linguistic expressions. This is the Referential Theory of meaning.
But, what is a referent? Consider a proper noun like “David Beckham”, what does it mean? It doesn’t
really mean anything. “David Beckham” is just a linguistic label which we apply to an individual. We
say that that individual is the referent of the expression.

But not all nouns are proper nouns. What about common nouns like table, chair, teacher, student,
love, peace.... They cannot have just one referent as the proper noun “David Beckham”. Common
nouns refer to sets of thins / individuals. Ex: “Table” refers to all those entities which can be
felicitously called like that. This set of referents is called extension.

The Referential Theory is a theory of meaning which establishes a relationship between language
and the world it designates. It is a simple theory, but it has many problems:

1. Function words (of, the) do not refer to anything. But as they do not denote, they fall outside the
interests of lexical semantics. What about abstract words (love, peace, etc....)? If we assume that
they exist in any relevant sense, Referential Theory does account for them, but language can name
things which do not exist for sure (for example, talking about the future or hypothetical events, also
imaginary elements like goblins or the Tooth Fairy, or words like happy cucumber.). The extension of
all these is an empty or null set: {}. But of course, we feel they mean something.

2. The biggest problem is that two expressions may have the same referent but mean different things.
Ex: A) Player number 8 in Barcelona FC.

B) Player who scored in the World Cup final.

Both expressions have the same referent: Iniesta. But it is possible for a speaker to know that he
plays for Barcelona FC and ignore that he scored in the World Cup final. Referential Theory cannot
explain this.

The solution was proposed by German philosopher Gottlob Frege. He introduced an abstract notion:
The sense, which mediates between language and the world. The sense is also called intension.

Sense: An abstract representation of the qualities that a referent needs to have so that we can apply
a label to it. In other words, the sense defines the information we need to refer. It is difficult to identify
an entity if you do not know what properties it has.

Law of Denotation: The more conditions the sense contains, the smaller the extension. Ex: Poodle
contains all the features of dog plus at least another one. Its sense is “bigger”, so its extension is
smaller (there are fewer poodles than dogs).

The separation of senses and referents explains why two expressions with the same referent (like the
case of Iniesta) can mean different things, because they have different senses. It opens the possibility
that languages have:

33
1. Expressions with sense but no referent.

2. Expressions with no sense but referent

3. Expressions with sense and referent.

4. Expressions with no sense and no referent.

Senses are what we are interested in Lexical Semantics. They correspond to the folk view on
meaning. But what are sense in the Mental Lexicon actually made of? They cannot be images
(remember what we said about images in lexical entries). Another possibility is that sense are
concepts.

Sense is a linguistic entity (part of lexical competence and lexical entries). Concept is an
extralinguistic notion (part of general knowledge). However, the line between linguistic and non-
linguistic meaning is very thin.

In contemporary linguistics: 1. Some believe that sense and concepts are the same (cognitivists. 2.
Some believe that senses and concepts are different entities. 3. Some take option 2, bit believe that
we also need concepts to construct meaning.

34
 15. The Classical Theory: Necessary and sufficient conditions.
Since classical times, the meaning of words is considered to consist of smaller parts. Ex: Girl:
Human, female and child compose this word. These sublexical elements are called semantic
components (or features / primitives). A lexical analysis which follows that strategy is called
compositional or componential.

Remember the meaning of “compositional” (lesson nº 2). Lexical theories which make use of
components are also called decompositional. The basic idea behind them is that senses can be
defined in such a way that they can be distinguished from other denoting units.

Consider the following definition for actress: “A woman or girl who acts in plays, movies or television”.
This definition contains two conditions which determine the reference:

1. An actress is a woman or a girl.

2. An actress acts in plays, movies or television.

These are necessary and sufficient conditions. They are necessary because neither can be left out,
and they are sufficient because they are all that are needed to distinguish actresses from non-
actresses. Note that the definition does not list incidental features: Famous, beautiful, etc....; so,
classical theory makes a clear distinction between definitional properties and encyclopaedic
information.

Definitional features contribute to the world’s sense. They are linguistic, which means that they are
part of the mental lexicon. They may activate information about a concept, but this belongs to world
knowledge / encyclopaedia. But of course, the problem is where to draw the boundary between
lexical and non-lexical meaning.

How do we describe word senses in a componential analysis? We make use of metalanguages:

Language-independent systems. They are language-independent because we don’t want to use
English to describe English and we want them to be valid for all languages (universal).

A metalanguage has a vocabulary and a grammar. The vocabulary is the semantic components.
Components which cannot be further decomposed are called primitives. Those are the building
blocks of meaning (for all languages). Example:

Woman: [FEMALE, HUMAN, ADULT]

Girl: [FEMALE, HUMAN, CHILD]

But if we want to make the features more versatile, we can make them binary: + / -. Example:

Girl: [+FEMALE, +HUMAN, -ADULT]

This reduces the number of components needed.

Important: Although we describe components with English word-forms in capital letters, they are not
English features or English concepts. You should understand them as universal meaning

35
components. We can use the components +FEMALE, +HUMAN, +ADULT to describe Spanish mujer
or German frau.

Note that we can also use the [A : V] format to state lexical meanings. Ex:

Woman: SPECIES: Human.

SEX: Female.

ADULT: +

This is common practice in many lexical theories. Another property of componential theories is the
use of redundancy rules. Ex: For dog (hierarchical structure):

Physical object.

Living thing.

Animal.

Mammal.

Beginning: If “X” is canine, then it is...

They are based upon hierarchical relations as in taxonomies. A given semantic component entails or
inherits other features. This simplifies representations.

Interim summary: Main features of componential approaches:

1. Complex meanings are reduced to a finite set of features.

2. These components can be applied to any language.

3. The define denotative meaning.

4. They can create an infinite number of meanings.

5. They express semantic relations among senses.

6. Redundancy rules express entailment relations.

An early componential approach: Katz. It was developed by linguists Katz, Fodor and Postal in the
60s. This was the first serious attempt to integrate semantics in generative grammar. The definitions
can be divided into two parts:

1. Genus: Semantic markers. ()

2. Differentiae: Distinguishers. []

Actress: (a woman or girl) [who acts in plays, movies or television]

Only semantic markers are responsible for the semantic relations among words. Ex: Actress: (a
woman or a girl) = HUMAN, FEMALE. So, an actress belongs into the category of HUMAN, FEMALE

36
entities and therefore holds a relation with other words like “lady”, “policewoman”, etc....; but cannot
refer to the same entity as a “man” or “filly”, because they do not share all markers.

An important contribution of Katz’s approach is the notion of selectional restrictions. These serve to
explain which combinations of lexical items are reasonable from a semantic point of view. Compare
*happy cucumber or * wary sausages (happy and wary go with animate or sentient entities). We put
selection restrictions in our lexical entries with angle brackets “< >”.

Although Katz’s theory has been very influential. It has a number of problems:

1. It’s been proved that some distinguishers are needed to explain word use.

2. The distinction definitional / encyclopaedic features is difficult.

37
 16. Fuzzy categories. Rosch and the Prototype Theory.
Basic premises.

1. Word meanings are not usually as precise as “necessary and sufficient conditions” make them out
to be.

2. Meanings are often imprecise and context-specific.

In other words: Word meanings may have fuzzy boundaries, meaning that a word’s denotation may
be hard to specify exactly.

Labov’s experiment.

Give the name of these objects, (1), (2) and (3):

(1) is a cup, while (2) is unknown for us and (3) is a bowl or vase.

Items that are difficult to categorise, like (2), might be categorised differently
in different contexts. Ex: If that element has coffee, people will tend to call it
a cup, but if there is ice cream in it, it will be probably called a bowl.

If (1) had ice cream in it, you might very well call it a cup with ice cream in it, demonstrating that both
the shape of something and its function determine whether it is a cup or a bowl.

In the classical tradition, the fact that ice cream usually goes in bowls is considered encyclopaedic
information that has nothing to do with the meaning of bowl. Labov’s experiment hints that such
encyclopaedic information cannot be dismissed as irrelevant to denotation.

Prototype Theory.

It provides an explanation for the way word meanings are organised in the mind. It is argued that
words are categorised on the basis of a whole range of typical features.

Prototype theory has its philosophical roots in Wittgenstein’s famous remark that the things covered
by a term often share a family resemblance, and it has its psychological roots in Eleanor Rosch’s
experimental treatment of much the same idea.

Rosch suggested that, when people categorise items, they match them against the prototype, or
“ideal exemplar”, which contains the most representative features inside the category.

Objects that do not share all the characteristics of the prototype are still members of the category, but
not prototypical ones. She argued that prototypes represent a “basic level of categorization”, e.g.
chair, as opposed to a superordinate, e.g. furniture and a subordinate level, e.g. kitchen chair.

38
In this theory, there is a scale of (proto)typicality, which goes rom more typical to less typical or more
peripheral. Examples:

Experiment 1: What is the best example of the category furniture? After surveying 200 American
college students, the resulting rank shows:

1. Chair, 2. Sofa, 3. Couch, 4. Table, 5. Dresser, 6. Rocking chair, 7. Coffee table, 8. Easy chair....

And the last members were:

.....55. Pillow, 56. Wastebasket, 57. Sewing machine, 58. Stove, 59. Refrigerator and 60. Telephone.

Experiment 2: What is the best example of the category bird? Take a look at this diagram:

Ostrich (less prototypical)

Penguin

Peacock

Owl, toucan

Dove, canary

Robin (+ prot.)

Sparrow

Parrot, pheasant

Duck

It clearly shows that native English speakers consider the robin to be the best example of a bird, and
the penguin and the ostrich as the worst examples of the category,

Evidence came from experiments involving:

1. Response times: People are quicker to recognise more prototypical members of a category as
being members of that category.

2. Priority in lists: Prototypical examples are among the first that people list.

3. Frequency: When asked to name a few exemplars, the more prototypical items came up more
frequently.

Prototypicality has a clear cultural component. For example, English speakers believe that robins are
prototypical birds, but Spanish speakers may think differently (gorrión, canario...).

Features can be core or peripheral. Core features are necessary conditions for belonging to a
category, while peripheral features are other features that some members of the category show.

39
Ex: Bird.

Core features: [Being feathered], [having two legs].

Peripheral features: [being able to fly], [being able to sing], [living in trees], etc....

But: What happens if we find a bird with no feathers? Is it still a bird? Or should we call it with another
name?

If we still call it a bird, the [being feathered] is no longer a core feature, but a peripheral feature. In
fact, this problem raises the question of whether we can really say with any certainty that any features
are core features.

Prototype theory has shown that definitional (linguistic semantic) and encyclopaedic (world
knowledge) aspects of meaning cannot be easily separated. Warning: Not all meanings have fuzzy
boundaries and not all prototype experiments have proved to be reliable. Think of the meaning of:
Hypotenuse or odd (odd number), there is not any fuzzy meaning or a scale of prototypicality.

40
 17. Componential analyses of lexical meaning I. Jackendoff’s
Conceptual Semantics.
Prototype theory has shown that the relation between lexical meaning and cognition is gradual. We
will now look at current componential theories of lexical meaning which are concerned with the
relation language – cognition. We will begin with Ray Jackendoff’s Conceptual Semantics.

For Jackendoff, semantic components are conceptual. Lexical concepts constitute a subset of all
concepts, that is, those which are lexicalized. For Jackendoff, lexical semantics is the study of how
lexical concepts are structured and how they interact with other language components (phonology,
syntax).

The lexicon in Jackendoff’s theory is the interface of three structures: Phonological, Syntactic,
Semantic, which are related with each other in a triangular structure.

Phonological, Syntactic and Semantics structures can interact in various ways. Consider the following
expressions: Press the cloth flat, paint the wall red, drink yourself silly. They all have the same
syntactic structure: [V NP Adjective], and they share a semantic component: Resultative.

As a componential theory, CS has a set of primitives and a grammar. The primitives include a number
of ontological categories, that is, types of entities that exist in the world and which we can designate
through language: Event, state, place, amount, thing, property, path, etc.... These are universal
semantic primitives.

Examples of ontological categories:

Ontology “is the philosophical field that attempts to organise everything that exists into a limited
number of general categories”. Words denote ontological categories. We can then generalize about
the categories they designate and study their structures.

There seems to be a correspondence between semantic categories ad word classes: Nouns typically
denotes THINGS, verbs typically denote situations (EVENTS, STATES) and adjectives typically
denote PROPERTIES. But the correspondence is not perfect. Examples: War, destruction, beauty,
intelligence, etc...

41
A semantic analysis in CS starts from one of these categories, which is elaborated with more
metalanguage. Consider the analysis of the following motion EVENTS: Pewee goes into the
clubhouse / Pewee enters the clubhouse. In each sentence we have a verb and two semantic
participants which complete the EVENT’s meaning. We call these participants arguments.

These arguments are semantically (and usually syntactically too) obligatory. Motion EVENTS are
represented in CS with a semantic component GO. GO has two arguments, a THING and a PATH,
and the whole structure is represented as an EVENT. Which ends up like this: [ EVENT GO ([THING],
[PATH])]

Note that GO is a semantic (conceptual) component, NOT the English word “go”. It is part of the
meaning of many motion verbs: Enter, walk, run, cross, etc... So, this is thus the general template of
motion verbs, but then we can elaborate the structure further analysing both PATH and PLACE.

We can further analyse the PATH argument: [ PATH TO [(PLACE)]]

And then the PLACE: [ PLACE IN [(THING)]]

We now have all the semantic components that we need to analyse the sentences. Using the
previous example:

[EVENT GO ([THING Peewee], [ PATH TO ([PLACE IN ([THING the clubhouse])])])]

Note that this structure can also be represented with a tree diagram. For example (not the same
example as before):

We can now represent senses in bracketing notation:

Enter: [ EVENT GO ([THING], [ PATH TO ([PLACE IN ([THING])])])]

Go: [EVENT GO ([THING], [PATH])]

Into: [ PATH TO ([PLACE IN ([THING

One problem with this notation, though is that it is difficult to read. In our entries, then, we will
separate the event structure:

Enter: GO ([THING], [TO ([IN ([THING])])])

Event1 (x, y)

42
Now, if you remember the SEM box contains a feature EVENT STRUCTURE. This goes in there.
This indicates that the sense of the verb enter is made up of one EVENT which takes two arguments.
Note that we can also have more than one EVENT and even different types (State, Property, etc...).

43
 18. Componential analyses of meaning II. Pustejovsky’s Generative
Lexicon.
James Pustejovsky: The Generative Lexicon (GL)

- GL is focused on the interface between lexical and phrasal meaning.

- GL shares many of the assumptions of Jackendoff’s work.

- GL assumes that meaning is conceptual in nature.

- GL has its origins in computational linguistics.

- Fuzzy meaning and prototype effects have not been dealt with in GL.

- GL helps to account for how we “fill in” a lot of information in context.

In this theory, there are four levels of semantic structure:

1. Event Structure: Relevant to verbs (state, process, transition).

2. Argument Structure: How many and what type of other expressions are required by the word.

Ex: Drive requires a driver (x) who needs to be an animate individual, and a vehicle to be driven (y).

Drive

ARGSTR. = ARG 1 = x: Animate individual.

ARG 2 = y: Vehicle

In many approaches, linguists generalize over types of arguments. They say that arguments receive
a semantic role (also called Theta-role). Typical examples are Agent and Patient in transitive Action
verbs. So, in our example:

Arg1: Agent (substitutes x) <animate individual>

Arg2: Patient (substitutes y) <vehicle>

3. Lexical Inheritance Structure: How the lexeme fits into the hierarchical structure of the lexicon (a
robin is a TYPE of bird). In this theory, Lexical Inheritance information is considered background
information (not included in the Attribute-Value Matrix)

4. Qualia Structure: Essential attributes of word’s denotata. Four Quales:

A. Constitutive: The relation between an object and its constituents or proper parts, which are the
material, the weight and the parts and component elements.

B. Formal: It identifies the element that distinguishes the denotatum form other things in a larger
domain, which can include size, shape, colour and position.

44
C. Telic: Purpose and function of the object, which is the purpose that an agent has in performing an
act and the built-in function or aim which specifies certaion activities.

D. Agent(ive): Factors involved in the origin or “bringing about” of an object, being this the creator,
the artifact, the natural king and the causal chain.

Ex: A novel.

- A novel (X) consists of narrative.

- A novel (X) is a book.

- A novel (X) is for someone (Y) to read.

. A novel (X) “is born / created” when someone (Z) writes it.

Which in the graph would be:

Novel

QUALIA = Constitutive = Narrative (X)

Formal = Book (X)

Telic = Read (Y, X)

Agent = Write (Z, X)

“Phil just finished a novel”: There are two interpretations: ....finished reading a novel (telic) / ...finished
writing a novel (agentive). However, nobody would interpret it as: ....finished *eating / *fixing /
*hanging a novel.

“Phil just finished lunch”: There are two interpretations: ....finished eating lunch / ...finished preparing /
cooking lunch, but never ....finished *reading / *writing lunch.

In fact, these examples show a phenomenon called type coercion, which is a semantic transformation
in which context-particular senses of words arise through the interaction of semantic structures in
phrases.

In these examples, the verb finish requires a person (X) and an event (E) (someone who finishes
doing something.). Since book or lunch are not events, the event interpretation is coerced from the
entries for book or lunch, thus obtaining the correct interpretation.

Words may have more than one sense. For example, polysemous words with systematic alternations.
Ex: Container / Containee: Mary broke the bottle / The baby finished the bottle.

Container (Object) / Containee (Information): That book is really thick / I disagree with that book.

45
19. Atomistic analyses of lexical meaning: Fodor and the postulates of
meaning.
The approaches to lexical meaning that we have examined (classical theory, Jackendoff,
Pustejovsky) are all compositional. We will now consider an alternative approach which argues that
lexical meaning is non-compositional or atomistic. This view is defended by philosopher Jerry A.
Fodor and it’s called Conceptual Atomism.

Conceptual Atomism claims that lexical concepts are primitive and have no structure. This may sound
counterintuitive at first sight. Entailment relations seem to prove that meanings can be decomposed.
Ex: If X is a bachelor, the X is an unmarried man. / If X kills Y, then Y dies.

So, what arguments are there in favour of the atomistic view on lexical meaning?

1. Necessary and sufficient features are difficult to isolate.

2. Typical features are defeasible.

3. Speakers do not always agree on lexical definitions.

4. Speakers do not necessarily possess all the essential features of a given sense. For example, you
need not to know that a whale is a mammal.

But the most important arguments are:

5. Since compositional approaches also use primitives, they also assume that some lexical concepts
cannot be decomposed.

6. If meanings break down into components, those with more components should take longer to
process. But this is not true!

How does Fodor explain the relation between senses and referents? He claims that all that is needed
is a reliable system that connects lexical concepts with the things they refer to. He assumes that
concepts carry information in terms of valid inferences, which are represented by means of
implications called meaning postulates.

Meaning postulates are not necessary and sufficient conditions: Some are necessary, some are
sufficient and some are typical. MPs are explanations, not definitions. Meaning postulates describe
relationships among concepts. Ex: If X is a chair, then it has four legs. / If X is a teacher, then people
learn from X. / If X is crimson, then X is red.

So, Fodor’s theory solves some of the problems we have detected in compositional theories, but it
has a number of problems itself:

1. It cannot generalize over classes of lexemes (motion events, properties, etc...).

2. Since concepts are not compositional, the theory assumes that we cannot construct new concepts
compositionally.

46
3. It assumes that all concepts must be innate, even for items like carburator, radiotherapy, etc....

Therefore, we will not consider Fodor’s model in this course, but you should remember that it is an
interesting alternative to compositional approaches.

47
 20. SEM content in lexical entries.
Even though that we have an IMAG. box, it is impossible for us to use it, so for practical reasons we
will leave it out. The same happens with the DEN box. All this means that the only categories or
boxes that interest us in the SEM content of lexical entries are the DEF box, the EVENT STR. Box
and the TH box(es).

For senses, we shall be making use of Jackendoff’s representations, but since they are visually
complex, we will separate the semantic components from the Event Structure. Using enter as our
example:

Enter: GO ([THING], [TO ([IN ([THING])])])

We can observe that there are two constants that interest us: The two THINGS (X 1, Y1). With this, the
SEM box would be the following one:

LEX: Enter

SEM: [DEF: GO ([THING], [TO ([IN ([THING])])])]

[EVENT STR.: EVENT1 (X1, Y2)]

[TH (1): Agent]

[TH (2): Locative]

(For Theta roles, the TH boxes, look at the Theta roles list)

In Theta roles, there are some interesting cases. For example, the passive affects Theme / Patients,
but not (easily) other Objects. Ex:

John killed Harry.

Harry was killed.

John resembles his father.

?His father is resembled.

“His father” is not then a Theme / Patient.

48
21. Semantic structure of lexemes which denote situations (events and
states) I.
Distinction Predicate / Argument

In grammar, sentences are organized around a main verb, but in semantics, propositions are
organized around its main predicate.

Sentences represent SITUATIONS (either STATES or EVENTS). A verb itself usually does not
express a complete SITUATION, so the verb interacts semantically with other parts of the sentence,
their arguments, to represent complete SITUATIONS.

Predicates / Arguments.

Ex: Fran picked the flowers. (SITUATION / EVENT / ACTION. Something is happening.) The
PICKING situation requires a “picker” and “something that is being picked”.

Predicate: PICK

Arguments: Two (the “picker” and “something that is being picked”; Agent and Patient)

Arguments have to be referring expressions (i.e., expressions that refer to things). Expressions like
“it” in sentences such as “it rains” would not count, as they do not refer to an element of the real
world.

Predicate: Pick

Arguments: Fran (Ag.), the flowers (Pat.).

Structure: Predicate (argument 1, argument 2) = Pick (Fran, the flowers) (Semantic analysis, it
represents a situation, but it is not a sentence).

But, not anything can be the “picker”, and not anything can be the “picked”. Ex: #Pictures picked
impishness. This sentence is grammatical, but semantically odd (thus #). The “picker” must be
something that can move (Fran can move, but pictures no), and the “picked” must be a physical entity
that can be grasped (flowers can be grasped, but impishness no). It is necessary to mark the
conditions of the NP.

The same happens with other predicates, such as erase or lift. These involve physical actions, and
require two arguments, but not any two arguments, which means that:

Argument 1: Must be a material thing that moves (Elvis or Eliza can be, but no flowers or the stone). It
is indicated with: Agent <animate>. Ex: Elvis erased the picture. / #Flowers erased impishness.

Argument 2: Must be a material thing, which is acted upon the moving thing (the picture or the stone
can be, but not impishness or calm), which would be written as: Patient <concrete>. Ex: Eliza lifted
the stone. / #The stone lifted calm.

49
Be careful: Syntactic structure is not always parallel to semantic structure. Consider the verbs eat,
draw, shave, washing, etc....

Semantic structure: Always two arguments (Agent, Patient): Eat (“Eater”, “thing eaten”), draw
(“drawer”, “thing drawn”), shave (“shaver”, “thing shaved”), wash (“washer”, “thing washed”).

Syntactic structure: Even if these are transitive verbs, they can be used intransitively (leaving out one
of the arguments, normally the patient one). Ex: I’ve eaten a meal already / I’ve eaten already, Jane
drew a picture in her notebook / Jane drew in her note. This phenomenon is called “unexpressed
object alternation”.

WARNING: This phenomenon cannot be extended to all transitive verbs; it depends on individual
verbs (or subgroups of verbs). For example, verbs having to do with caring for the hair cannot
undergo unexpressed object alternation. Ex: *Rakesh combed after his shower / Rakesh combed his
hair after his shower, *Karen curled on her wedding day / Karen curled her hair on her wedding day.
The sentences with “*” mean that they are syntactically wrong.

Difference between STATES and EVENTS in SITUATIONS:

- States: Stative verbs, more constant across time, unchanging.

- Events: Dynamic verbs, less constant across time, happening or changing.

Test for EVENTS: What happened? Ex: In the sentence “the clock stood in the hall” you cannot ask
what happened?, as there is not movement, but in the sentence “the children stood up”, you can ask
what happened?, which would be the same sentence, being it the EVENT. In fact, the part of “up” in
“stood up” is what denotes movement.

EVENTS / STATES = MOTION verbs / LOCATION verbs. Sheryl went into town on foot, Wilma left
(EVENTS). / The monument stands beside the fountain (STATE).

Event MOTION verbs denote changes in location, they are dynamic verbs that contribute to an
EVENT description. In Jackendoff: [ EVENTGO ([THING] , [PATH])].

State LOCATION verbs are stative verbs that locate something, not involving a change in location. In
Jackendoff: [ STATEBEloc ([THING], [PLACE])].

50
22. Semantic structure of lexemes which denote situations (events and
states) II.
Causation.

Many verbs indicate that something caused an EVENT or STATE to come about. Some MOTION
verbs are inherently causative. Ex: Raise. “Peter raised the flag” in fact means that “Peter caused the
flag to move upwards”.

Other examples: “The ball rolled down the hill” (Not causative) / “Sally rolled the ball down the hill”
(Causative = “Sally caused the ball to roll down the hill”). It is necessary that there are always two NP
for a verb to be causative, one is the element that causes the action (Sally), and the other is the
element that does the action (the ball).

Causative alternation.

Roll1 = 1 argument (The ball, this is the one of the verb roll as an intransitive verb). In the formula, it
would be expressed as follows: Roll1 = [EVENTGO + Roll ([THING], [PATH])]

Roll2 = 2 arguments (Sally, the EVENT of the ball rolling. this is the one of the verb roll as transitive
verb, being also causative). In the formula, it would be expressed as follows: Roll2 = [EVENTCAUSE
([THING], [ EVENTGO + Roll ([THING], [PATH])])]

The complete analysis of the sentence “The ball rolled down the hill” would be as follows:

[EVENTGO + Roll ([ THINGthe ball], [ PATHDOWN ([ THINGthe hill])])]

And the complete analysis of the sentence “Sally rolled the ball down the hill” would be as follows:

[EVENTCAUSE ([THINGSally], [EVENTGO + Roll ([ THINGthe ball], [ PATHDOWN ([ THINGthe hill])])])]

Conflation.

Consider these examples:

A. James entered / went into the room by walking.

B. James walked into the room.

Both are correct in English, but (B) is more natural than (A), which happens because English tends to
naturally conflate MANNER with MOTION EVENTS. In the analysis, it would be written as follows:
[EVENTGO <walked> ([ THINGJAMES], [ PATHinto the room])]

Extending the LOCATION / MOTION structures.

- STATE: [Beloc ([THING], [PLACE])]. But for other cases, PLACE would need to have the following
elements, depending on the case: BEident = PROPERTY / BEposs = Possessor THING / BEtemp = TIME.
Examples:

+ Charles is giddy: [ STATEBEident ([THINGCharles], [PLACEAT ([PROPgiddy])])].

51
+ Hugh has a pencil collection: [STATEBEposs ([THINGa pencil collection], [ PLACEAT ([THINGHugh])])].

+ The seminar is at 5pm: [STATEBEtemp ([THINGthe seminar], [ PLACEAT ([TIME5pm])])].

- EVENT: The GO component can be extended in the same way. All of the following are types of
EVENT. Examples:

+ Charles went giddy: [ EVENTGOident ([THINGCharles], [PATHTO ([PLACEAT ([PROPgiddy])])])].

+ Hugh’s pencil collection went to his heirs: [ EVENTGOposs ([THINGHugh’s pencil collection],
[PATHTO([PLACEAT([THINGHugh’s heirs])])])].

+ The seminar went from 2pm to 5pm: [ EVENTGOtemp ([THINGthe seminar], [ PATHFROM ([PLACEAT
([TIME2pm])]) TO ([ PLACEAT ([TIME5pm])])])]

+ The seminar lasted until 5pm: [ EVENTGOtemp ([THINGthe seminar]. [ PATHTO ([PLACEAT ([TIME5pm])])])].

52
SEM 0

Competence (knowledge of language)/performance

1. Is the student competent in his use of language? Yes, he knows english, they are english words, independent or
what type
2. What about his performance from a social viewpoint? He doesn’t’ adhere to a standard, in this context he does
not go with the social norms of this interaction
3. Do the student and the teacher have a similar linguistic competence? Yes, both speak english, they both have a
knowledge of the language
4. Do they have the same lexicon?(if they do not, whose lexicon should lexicology study, the student or the
teacher’s) yes, even though the vocabulary is different. The mental lexicon is the same. Vocabulary are the
real expressed words, while lexicon is the words we know and understand

Lexicology/lexicography
1. Should a dictionary include the spelling variants used by the student? Yes, if the meaning is unpredictable.
2. What about expressions like good yard or lol? Yes, if it ends up being a common expression in english
3. If these expressions are bad language, what should lexicology do about them? Also yes, ‘bad language’ is a
part of the description,
Key issue: are the student and the teacher speaking the same language? Yes, they display different vocabulary, but
using the same language.
Rebuses or rebus writing: used in social contexts, usually from the young people, abbreviations and expressions

SEM 1

Cambridge dictionary: “fine”

Adjective: good or good enough; healthy and well
Adjective: excellent or much better than average
Adjective: very thin or in very small pieces or drops
Adjective: very exact and delicate, or needing to be done, treated, or
considered very carefully
Adjective: sunny and dry (mainly in UK)
Adjective: bad or not convenient
Noun: an amount of money that has to be paid as a punishment for not
obeying a rule or law
Verb: to charge someone an amount of money as a punishment for not
obeying a rule or law

Oxford dictionary: “fine”

Adjective: very well
Adjective: acceptable/good enough
Used to say you are satisfied with something
Adjective: attractive
Adjective: very thin

Cambridge dictionary: “drink”

Noun: (an amount of) liquid that is taken into the body through the mouth
Noun: alcoholic liquid
Verb: to take liquid into the body through the mouth
Verb: to drink alcohol

Oxford dictionary: “drink”

Noun: a liquid for drinking; an amount of a liquid that you drink
Noun: alcohol or an alcoholic drink; something that you drink on a social occasion
Noun: a social occasion where you have alcoholic drinks
SEM 2
words:
- culture: (noun) content word. Set of characteristics that join
a group of people/ knowledge or ideologies from a specific
group of people/ common knowledge
- oak: (noun) content word. Type of tree that only grows in
humid climates that produces acorns/ a taste given to older
wine from storaging it in oak barrels.
- google: (noun) content word. Tool used to search in the
internet/ multimillionaire company/ the act of searching
online.
- the: article, function word. Used when connecting nouns to the rest of the sentence/ used to define importance to
the specific noun we are talking about
- efficient: adjective, content word. Used to refer to a person that manages time and resources in the best way
possible/ used to refer to an object that simplifies or facilities an action

words

grammar lexical
(= function) (= content)

SEM 3
Consider the following expressions and decide on the number of words and lexemes they contain. You should apply
Murphy’s criteria as presented in the book.

Is the number of lexemes different from the number of words?

- Homework: 1 wordform (no blank space), 1
lexeme. It is a compound word, formed by 2 words
but with a single concept. The meaning of
homework is kind of ‘transparent’, home is the
element that deviates the most from the concept,
that is why is partially transparent and opaque.

- Four-wheel-drive (vehicle): 1 wordforms (no blank space, high fence compound), 3 lexemes, as each of the words
has its own meaning. Drive is not 100% transparent, it means that (the 4 wheels) gets power from the engine of the
car, noun that means power or energy in this context
Couldn’t: 2 words, 2 lexemes. Contraction is a phonological phenomenon, which has a spelling translation.
Give up: 2 word, 1 lexeme. Phrasal verb, not the combination of the 2 words, it is a full words that moves with its own
accent.
Ouch!: (interjection) 1 words, 0 lexeme. Doesn’t have a syntactic category, it won’t be part of a sentence, whenever
you say and interjection you put a comma. It is a special element, problematic in language. If we say it is a lexeme it
means that they have a meaning. It is a function element.
(You are) the apple of my eye: 5 words, 1 lexeme. Idiomatic expression
Fan-bloody-tastic: 2 words, 1 lexeme.
SEM 4
The following lexical items are ambiguous between (at least) two different senses. On the basis of Murphy’s criteria
(definition, contrast and specially zeugma), decide whether they illustrate cases of homonymy or polysemy in the
mental lexicon.
- Cousin: 1. ‘a female cousin’. 2.’a male cousin’
I have both a cousin and a cousin
I have both a female cousin and a male cousin
Homonymy
- Flyer: 1. ‘a pilot’. 2. ‘a small handbill announcing an event’
I saw a flyer in a flyer
The flyer entered the plane while carrying it
Polysemy
- Mouth: 1. ‘place where a river enters the sea’. 2. ‘opening in the lower face’
- Decline: 1. ‘reject’. 2. ‘conjugate a grammatical word’
- High: 1. ‘affected by drugs’. 2. ‘significantly above the ground’

*Pedro ejecutó la orden y el comandante a los prisioneros

*Le echó de menos y las cartas
*Tomó un vaso de agua y las de Villadiego
*Llegó a la cumbre de la montaña y el presidente a las del estado

TASK 1

Cousin:
- Synonym: there is no lexeme available
- Antonym: niece
Flyer:
- Synonym: throwaway, pamphlet
- Antonym: cabin attendant
Mouth:
- Synonym: oral cabity
- Antonym: eyes (upper cabity), nasal cabity
Decline:
- Synonym: refuse, turn down
- Antonym: accept
High:
- Synonym: stoned, tall
- Antonym: low

TASK 2
TASK 3

SEM 6

SEM 7
The following are examples of attested speech errors. Classify them as semantic and / or phonological (malapropisms)
and describe the cause of the error. What do they tell you about the relation of items in the mental lexicon? What
lexical relations have motivated the errors?

Irvine is quite clear (close/near): phonological. The relationship of the words is synonymy, and when you put them
together makes an already existing word, there is a confusing relation between the intended word and the one said
To determine watch (which/what): phonological, they are both ‘wh’ words, but the problem is the ‘blending’ of both
pronunciations
At the end of today’s lection: phonological. This is an example of ‘blending’, putting together
2 words (like breakfast and lunch for brunch), lection is a result of blending 2 words, ‘lecture’
and ‘lesson’, which have a relation in semantics, like both are types of classes.
He’s a high grader (low): semantic. The relationship between the words is antonymy
Don’t burn your toes (fingers): semantic. We use finger as a generic
I just put it in the oven at a very low speed: phonological. We use temperature instead of
speed, this is a question of colocations
You can cut rain in the tree: syntactic/semantic. The structure in the chain is the problem
I have to smoke my coffee with a cigarette: semantic. Same as the previous one

However, they delineate…quotas, I think, vulcanize society (George W. Bush): phonological, trying to say the
Balkans, the place of the countries, with the intention of referring to the process of dividing (the Balkans into different
countries) the society
I might just fade into Bolivian, you know what I mean? (Mike Tyson): phonological, ‘fade into oblivion’ was the
correct expression
SEM 8

(hoja)

SEM 11
Task 1. Give a neutral term for the following ‘formal’ words.
Hitherto, incumbent (adj.), inveracity, laudation, lavation, lubricious, manifold, mariner, natation, yesteryear.
Should they be marked in their lexical entries as [+formal] or should their equivalents be marked as [+neutral]?
Hitherto: yet
Incumbent (adj.): mandatory
Inveracity: lie
Laudation: praise
Lavation: wash
Lubricious: lewd
Manifold: many/several
Mariner: sailor
Natation: swimming
Yesteryear: past
Task 2. Classify the following as either British or American English words.
baby carriage - pram gasoline, petrol,
subway - tube (baby) pacifier - (baby) dummy
line - queue bathroom - lavatory
garbage - rubbish (car) hood - (car) bonnet
bookstore - bookshop apartment building - Block of flats
tap - faucet (pencil) eraser - (pencil) rubber

SEM 13
1. Which of the following statements express entailment relations? Treat X as placeholders for any noun phrase.
a. If X kills Y, then Y dies. No entailment
b. If X assassinates Y, then Y dies. entailment
c. If X shoots Y, then Y dies. No entailment
d. If X shoots Y then Y is hit by a bullet. No entailment
e. If X drowns, then X dies. Entailment
f. If X drowns, then X is submerged in water. No entailment
2. Determine whether the following statements express entailments. If so, what do they tell us about the
denotative meaning of teacher? If not, discuss where they express pragmatic inferences and whether they
reflect any connotation social meaning for teacher.
a. If X is a teacher, then X is paid to teach.
b. If X is a teacher, then X is female.
c. If X is a teacher, then people learn from X.
d. If X is a teacher, then X has taught.
e. If X is a teacher, then teaches.
Pagina 41 del libro
SEM 14
1. The distinction sense and referent opens the possibility that languages have:
a. Expressions with sense and referent
b. Expressions with no sense but referent
c. Expressions with sense but no referent
d. Expressions with no sense nor referent
2. For each of the following explain why it is necessary to distinguish the sense and referent of the expressions.
a. The chair I’m sitting on. the sense of chair ‘something you sit on’ allows you to identify the object in the
outside world, and there will be as many referents as many people say it, as many chairs people is sitting
on.
b. People who have been to mars. It has sense but does not have a referent
c. William Shakespeare. It has no sense, but does have a referent, same as ‘David Beckham’
d. The most famous English playwright. The referent could refer to multiple playwrights, but the sense needs
to be universal
e. The author of Romeo and Juliet. The referent and sense are not the same
SEM 15

Kill: necessary but the definition is not sufficient or

encyclopaedic
Tiger: encyclopaedic and sufficient
Doctor: not sufficient
Jog: Not necessary or encyclopaedic

SEM 16

Furniture:
1: chair
2: table, desk, bed
3: sofa, couch, easy chair
4: chest of drawers
5: dresser
6: bureau, coffee table
7: rocking chair

Bird:
1: robin
2: seagull, dove
3: sparrow, canary
4: blackbird, mockingbird
5: woodpecker
6: parakeet, swallow
7: lark, starling
SEM 17
The following CS represents the sentences which follow:

[ 𝐸𝑉𝐸𝑁𝑇 𝐺𝑂 ([ 𝑇𝐻𝐼𝑁𝐺 𝐴𝑚𝑒𝑙𝑖𝑎], [ 𝑃𝐴𝑇𝐻 𝐹𝑅𝑂𝑀([ 𝑃𝐿𝐴𝐶𝐸 𝐼𝑁([ 𝑇𝐻𝐼𝑁𝐺 𝑡ℎ𝑒 𝑝𝑙𝑎𝑛𝑒])])])]

1. Determine which part of the tree corresponds to the bold words in the sentences and rewrite the representation
in bracketed notation
- Amelia deplaned
- Amelia went out of the plane: [GO ([THING], [PATH])]
- Amelia exited the plane: [GO ([THING], [ 𝑃𝐴𝑇𝐻 𝐹𝑅𝑂𝑀([ 𝑃𝐿𝐴𝐶𝐸 𝐼𝑁 ([THING])])])]

2. Deplane can be both an intransitive and a transitive verb as in the troops were deplaned an hour later. It can
be paraphrased as ‘cause to leave an aircraft’. How would you modify the CS to account for this transitive
use?

The students surrounded the building

[LEX: surround]
PHONOLOGY: [SEGMENTS: səˈraʊnd]
[SYLL STR: syl(1): sə / syl (2): ˈraʊnd]
[STRESS: syl (2)]

SEMANTICS: [EVENT STR: EVENT (x1, y2)]

[TH (1): AGENT]
[TH (2): LOCATION]

SYNTAX: CAT: [MAJOR: verb]

[NUMBER: singular]

C-SELECTION: [COMPL (1): NP]

[COMPL (2): ]

MORPHOLOGY: [M-TYPE: stem]

[M-CAT]
[M-SELECTION]
[M-STR]
[HEAD]
[COMPL]
[PAST]
[PPT]
[PL]
[LEX]
SEM 18
Provide the qualia structures for the following words. Discuss the problems you have encountered

Lunch

(x) as the item and (y) as the person, (z) instigator

Dictionary
Constitutive: dictionary is made of lexical entries
Formal: dictionary is a type of book/ website/ CD-ROM
Telic: what you do with a dictionary is consult it
Agentive: dictionary comes into being when someone compiles definitions of words in the same document/place
Constitutive = lexical entries/ pictures (x)
QUALIA = Formal = book/ website/ CD-ROM (x)
Telic = consult (y, x)
Agent = compile (z, x)

Sandwich
Constitutive: sandwich is made of at least to pieces of bread,
Formal: sandwich is a type of lunch
Telic: what you do with a sandwich is eat it
Agentive: sandwich comes into being when someone success in preparing it
Constitutive = food (x)
QUALIA = Formal = lunch (x)
Telic = eat (y, x)
Agent = prepare (z, x)

Poem
Constitutive: poem is made of verses/ lines
Formal: poem is a type of stanza
Telic: what do you do with a poem is read it
Agentive: poem comes into being when someone writes it
Constitutive = food (x)
QUALIA = Formal = lunch (x)
Telic = eat (y, x)
Agent = prepare (z, x)